Amazement

I’m old. I’ve seen good times and bad. I’ve seen people at their best and their worst.

I’m absolutely amazed that a significant number of USAians (and visitors) in Florida are not evacuating from the east coast of Florida in the face of an F4 Hurricane (Matthew) with multiple weather-models predicting:

  • The eye of the storm will hit the shore near Melbourne and ride nearly along the shore for a couple of hundred miles of densely populated high-value property.
  • Storm surge is predicted to be greater than 9 feet high miles inland.
  • Waves of up to 20 feet on top of the storm surge and rain threatens flash flooding, destruction and death miles inland.
  • The path of destruction could be as bad as an EF 2 tornado but 25 times wider and 25 times longer and thousands of times more homes.
  • Power is expected to be out for many days for millions of residents.

In case these numbers mean nothing:

  • EF 2 tornadoes damage light buildings and destroy trees and powerlines. “Considerable damage. Roofs torn off well-constructed houses; foundations of frame homes shifted; mobile homes completely destroyed; large trees snapped or uprooted; light-object missiles generated; cars lifted off ground.” but usually cover only a few square miles. This thing will cover ~1K square miles.

So, I have to ask, “Do these people feel lucky?”

I’m amazed. Tolls have been cancelled, inbound lanes of highways reversed, supplies of fuel, school buses, National Guard, and ~100 substantial shelters have been set up and still folks don’t evacuate. The governor and many others have repeatedly stated this storm will kill people who don’t evacuate. Who will the survivors blame in the coming week when the dead are piled high in makeshift morgues? Will the only good to come out of these events be that a million Trump-supporters have died, that the gene-pool will have been sharpened for rationality traits, or that the economy will have a major boost repairing the damage?

About Robert Pogson

I am a retired teacher in Canada. I taught in the subject areas where I have worked for almost forty years: maths, physics, chemistry and computers. I love hunting, fishing, picking berries and mushrooms, too.
This entry was posted in politics, weather and tagged , , , . Bookmark the permalink.

176 Responses to Amazement

  1. oiaohm says:

    IRS definitions comes the IRC, which promulgate from the USC and CFR.
    Dougman you need to look again.
    “Internal Revenue Code”
    Applies to IRS when they are doing tax code. Not collating information from tax collected data to send to other department or trade overseas for compare data . There are three books of defines used by the IRS.

    That it is trade information is why the IRS uses three name for one define on collated data one going inside the USA and two are to be exported. Anything directly define in the IRC is only allowed one define. So putting collated data define for export and internal USA usage in IRC is not workable.

    Basically dougman you don’t even know where to look to find the define. Yet you though you knew enough to claim that USA would have to increase taxes to have a good health system. Collated data from the IRS is exported to Australia most of the OECD and many other countries. Those documents include defines for the words I used. Of course other tax departments don’t hand over their collated data without receiving collated data in return. This is the big wide world of statistics.

  2. dougman says:

    “No man up to being a complete idiot.”

    Actually, you’re the idiot here.

    “You quoted a hoax that is absolutely false.”

    So you think things that NASA say are false? What about the Earth going into darkness, is that false too?

    “Now you are attacking me over something that is absolutely true.”

    I am calling you out for being a fraud, Fifi. Lets get that straight. To date, you still have not defined, “Averaged American” or “average USA person” per the IRC. However, you confused yourself by your own ignorance by thinking I was referring to mIRC, which is laughable!

  3. dougman says:

    “Please note its not IRC and this shows you can not proof read IRS.”

    LOL!….IRC stands for Internal Revenue Code

    “IRS define comes from providing information United States Census”

    WTF?…you’re so lame. IRS definitions comes the IRC, which promulgate from the USC and CFR.

  4. oiaohm says:

    There is no such thing as “Averaged American” or “average USA person” per the IRC, you just pulled it out of your feeble empty head.
    Please note its not IRC and this shows you can not proof read IRS. IRS define comes from providing information United States Census. Same thing happens in most countries with a central Census collecting body.

    Come on, man up to your ignorance!
    No man up to being a complete idiot. You quoted a hoax that is absolutely false. Now you are attacking me over something that is absolutely true.

    Tax departments have to report data to Census body like everyone else in a country. Part of them reporting to the Census body is producing averages.

  5. dougman says:

    ” So that means I really not going to waste me time handing over my research to you.”

    LMAO, so you an admitted fraud. Something we all have known.

    There is no such thing as “Averaged American” or “average USA person” per the IRC, you just pulled it out of your feeble empty head.

    Same thing goes for, “Mind you the IRS for internal USA documents uses average US person instead of average USA person” and “”You should not have been asking me what those terms are if you had read the documents by the IRS on tax load as both appear in the IRS documents.”

    Come on, man up to your ignorance!

  6. oiaohm says:

    Here is an original though, perhaps they came from my own brain? Amazing, eh??
    So this statement says I don’t have to provide you with any cites because you will just making stuff up. I consider that it had come from you own brain and had based the idea on nothing in facts. So that means I really not going to waste me time handing over my research to you.

  7. dougman says:

    “Dougman you said all this crap without putting up any sources. So before I have put sources I want to see you sources to your crap. I have already put up enough sources that points a solid without explain the other bits.”

    Here is an original though, perhaps they came from my own brain? Amazing, eh??

    Still waiting on the following:

    “‘Averaged American’ is a term form Harvard USA.”

    Source?

    “‘average USA person’ is term from Internal Revenue Service.”

    Source?

    “Mind you the IRS for internal USA documents uses average US person instead of average USA person.”

    Source?

    “You should not have been asking me what those terms are if you had read the documents by the IRS on tax load as both appear in the IRS documents.”

    Source?

    Your move chump, or should I call you chimp, as you paint your face black?

  8. oiaohm says:

    Basically dougman has been caught in another bold face lie and is now attempt to nit pick way out.

  9. oiaohm says:

    To make this happen, government would have to increase taxes to every citizen, which can plunge United States into a socialism state where government controls the distribution of goods and services. United States is currently practicing free market system where market forces determine the availability and cost of health care services and not the federal government. The introduction of universal health care will lead to great distortion in the economy. It could be a first step towards government controlling other aspect of the economy.
    Dougman you said all this crap without putting up any sources. So before I have put sources I want to see you sources to your crap. I have already put up enough sources that points a solid without explain the other bits.

    You asking for sources is drival because you never had valid sources in the first place and made this garbage up or followed someone saying made up garbage.

  10. dougman says:

    Disregarding all your other drivel, as typical you failed to cite your sources per the IRC.

    “‘Averaged American’ is a term form Harvard USA.”

    Source?

    “‘average USA person’ is term from Internal Revenue Service.”

    Source?

    “Mind you the IRS for internal USA documents uses average US person instead of average USA person.”

    Source?

    “You should not have been asking me what those terms are if you had read the documents by the IRS on tax load as both appear in the IRS documents.”

    Source?

  11. oiaohm says:

    dougman take the time and go look up the percentage of GDP Australia USA and Canada government spends on health.

    If the percentage of government budget going into health aligned to the quality of health care system you got the USA should be kicking Australia and Canada ass. Year 2000 USA was spending less than 1 percent GDP on health today spending over 17 percent in the USA.
    Year 2000 Australia was spending over 17%GDP on health but today we are spending 10%.
    Year 2000 Canada was spending over 10% GDP on health but today under 6%.

    Here interesting point in the year 2000 USA Canada and Australia were all the same rating in health system. The two that have reduced costs have in fact improved the one that has just attempted to throw money at it has got worse. This has been the interesting thing most of the ones to move up in rating since the year 2000 have being decreasing costs. Australia is one of the most expensive to run that is in the top 30.

    http://www.theaustralian.com.au/business/australia-ranks-among-top-countries-in-global-efficient-healthcare-rating/news-story/660d04ad1b71ed02932d446f7d9c6a0a?nk=da3094d0f77929ecc161f4fc663bf485-1476928682
    This really brings the numbers home. For a heath system that is in the top ten Australian hand over average including what government pays $US6110. USA people getting treatment in what is classed as the 50 in the world system are handing over $US9146. So USA people are at least being gouged by 3000 dollars. So dougman you are paying almost 1/3 more for worse service.

    Do remember what I wrote before Australia is one of the most expensive in the top 30 and we are only $US6110. Australia has to import most of the drugs we use. USA makes most of the drugs that are used in the USA. So out of the $US6110 we are losing a slice to drugs and medical products lost at sea. So Australia due to where we are and what we have to import should be fairly expensive. USA on the other hand where you don’t have sea transport for most medical products should be fairly cheep.

    This is why all the talk that hey to have a better health system in the USA we have to increase taxes is bogus. The issue is the USA health system is totally not efficient. Just like the Epipen problem. When a medical services are provided there are to many people taking a slice. A prescription paper something doctors use lots of right. From production to doctor for a simple bit of prescription paper the average number of middle men is 12 in the USA. Canada and Australia for a bit of prescription paper the max is 4.

    If the USA mil run logistics like the USA civilian medical world does you would not have any money for anything other than the current wars you are in.

    The is the coward problem USA focuses on getting the stuff right to fight people but then fails to get other basic stuff right medical. I am not saying that USA has to implement a system exactly like Australia. For money spent USA should have a system that rates as well as Australia if not better.

    Here is the other shocking reality in per person spend on medical USA has higher spend value than the top 45 countries. So its not that the USA need to spend more money on health in fact USA needs to spend less on health but the money need to be spent effectively. USA has a lot of wasteful medical spending. In fact the USA has a lot of wasteful spending over all.

    There are a huge list of excuses out the USA but the value spent per person does not lie. It slows a complete efficiency problem. Hillary Clinton has mentioned a few thing that may help. So far Trump as just said lets just rip the system down without any of the regulations that need to be put in place to at least sort out the distribution system.

    The numbers say you should cap health spending at 11% of GDP as you really don’t get any more improvement spending more than that.

  12. oiaohm says:

    Robert Pogson the maths for tax are way simpler.

    Total tax collected for all taxs from the population divided by total income from the population gives a percentage.
    http://www.nowandfutures.com/taxes.html
    The USA average number is over 43%. Australia is 25.6 percent.

    Ok looking at Australian treasury numbers this can fool a USA person.
    http://www.treasury.gov.au/Policy-Topics/Taxation/Pocket-Guide-to-the-Australian-Tax-System/Pocket-Guide-to-the-Australian-Tax-System/Part-1
    We compare out total tax rate against the lowest taxed USA state not the USA as a total because it a total pain in the but adding up every USA states numbers. How do we make up the USA best possible state we take the huge wrong presume that USA states not apply any taxes. So Australian numbers show USA federal tax only put head to head with Australian total collected tax. This is also what the OECD does as well. Dougman how many USA states have no state taxs.

    The worst USA states make Denmark look good due to them have Tax burden over 60%.

    http://www.investopedia.com/financial-edge/0411/do-canadians-really-pay-more-taxes-than-americans.aspx
    Canada and USA for average tax is basically identical. So they are not really pay more tax or less tax but the same tax. Individual people will find they will pay less tax in the USA or in Canada but that is on a individual base and what USA state they happen to be in compared to Canada.

    If the amount of tax you paid and the quality of your health care system were linked USA and Canada should be kicking Australia ass.

  13. dougman says:

    Neither of you nincompoops, can cite me a source for the wording as proffered, evidence denied.

  14. dougman wrote, “Never seen them, and you just made them up.”

    The concept is rather simple. What kind of person would have the same effect on the national economy if everyone were identical? e.g. Suppose there are three USAians. One make $40K, another $50K and another $80K per annum. The “average USAian” would be a person making â…“ of ($40+$50+80)K = $56666.666. There might not be such an actual person but if there were three of them, they would have the same total income.

    The concept is very useful in such simple cases but often you have quite wide-spreading data with two or more clumps that the average is not particularly representative, in any way shape or form. e.g. Suppose the third USAian makes $800K. Then the average is not representative of any particular class of USAians although it still has the same total effect.

  15. oiaohm says:

    dougman you put up that USA had lower tax than Australia and Canada so before I show my sources you need to show yours.

    “Both mean exactly the same thing.”
    You say they don’t mean the same thing. They do mean the same thing when you are talking in terms of IRS documents.

    Sorry you raised this topic so you have to present you sources first then I will present my and rip you a new one.

  16. dougman says:

    “‘Averaged American’ is a term form Harvard USA.”

    Source?

    “‘average USA person’ is term from Internal Revenue Service.”

    Source?

    “Both mean exactly the same thing.”

    LOl, nope.

    “dougman so you should have known at least 1 of those two terms.”

    Never seen them, and you just made them up.

    “Mind you the IRS for internal USA documents uses average US person instead of average USA person.”

    Source?

    “You should not have been asking me what those terms are if you had read the documents by the IRS on tax load as both appear in the IRS documents.”

    Source?

    Just making up crap as you going along, eh?

  17. oiaohm says:

    https://www.ato.gov.au/Business/GST/When-to-charge-GST-%28and-when-not-to%29/GST-free-sales/
    Ham-Dweeb, trying to say that a Sales tax and a GST is one in the same laughable, as a GST is a much broader tax. GST taxes almost everything.
    Really GST Australia is two rates. 0 percent and 10 percent. GST free items don’t get taxed at all. Australian GST is not a pure flat rate GST its a two flat rates GST.

    https://en.wikipedia.org/wiki/Sales_taxes_in_the_United_States#Services
    USA sales tax depending where you are and what the service it may be sales taxed.

    Yes USA Sales tax is a form of good and services tax for what it covers. Just with a lot more different rates.

    GST is a tax on both goods and services at a flat rate of 10 percent. It is effectively a tax paid by the end user.
    That is a very key point. Unlike a sales tax the USA has a GST means that the GST on the parts that make up a item being sold that has already been paid is subtracted from the amount of GST that has to be paid by the person making the item. You don’t get tax multi stacking as you get in the USA sales tax system. %2.5 tax multi stacked can end up crossing %10.

    https://www.business.gov.au/info/run/tax/register-for-goods-and-services-tax-gst
    Australia is not that simple. Not all businesses are required to collect GST. Even if the government says X service will have GST on it does not mean it will.

    Do note the the key bit is a business is allowed to claim back the GST on everything the business buys for the operation of the business if they are GST registered.

    GST is a broad based consumption tax which is ultimately borne by the consumer.
    This is repeating the mistake of the prior over who pays GST. If a business is a consumer and they are GST registered the GST is not borne by them with how the Australian system works. Ultimately borne by a individual consumer not business consumers in most cases.

    You also said end user that is also wrong. Again if the end user is a GST registered business they are not paying GST on it in Australia and if they do they claim it back.

    A sales tax is borne by business True that is the problem. The final product costs are normally based on the cost of business to make profit. So sales tax hits business then business sets pricing to recovering costs including the sales tax they had to pay on the individual who is also being hit by sales tax again.

    When you have seen the number run by the USA IRS and the Australian tax department on tax load caused by sales tax vs GST the reality is Australian GST works out on average lower. Its purely due to the anti-stacking nature of GST.

    “Averaged American” is a term form Harvard USA.
    “average USA person” is term from Internal Revenue Service.
    Both mean exactly the same thing. dougman so you should have known at least 1 of those two terms. Mind you the IRS for internal USA documents uses average US person instead of average USA person. Yes a “us person” makes Australians laugh. You should not have been asking me what those terms are if you had read the documents by the IRS on tax load as both appear in the IRS documents.

    So basically you claiming USA has low tax rate by wild guess because again you have done no research into the topic before attempting to make point.

  18. dougman says:

    “Yes Australia has a GST. When you average it out…GST compared to what Averaged American pays in sales tax and surtax the result is average USA person pays more than what an Australian pays in GST.

    LOL..wtf is a “Averaged American” and an “average USA person”.

    Ham-Dweeb, trying to say that a Sales tax and a GST is one in the same laughable, as a GST is a much broader tax. GST taxes almost everything.

    The Sales Tax rate ranged from around two and a half percent to around thirty percent. It did not tax services.

    GST is a tax on both goods and services at a flat rate of 10 percent. It is effectively a tax paid by the end user.

    Sales Tax applies to goods at varying rates.

    GST applies to almost everything at a set rate.

    GST is a broad based consumption tax which is ultimately borne by the consumer. A sales tax is borne by business, etc

    But You being a “tax-authority” already supposedly knew this.

  19. oiaohm says:

    Robert Pogson I just re-looked at the numbers. The USA must be run by idiots. Australia thinks spending %10 of GDP on all health programs is excessive but the number 1 status Australia has does come with a price tag. Canada system is done on the cheaper side but still quite decent. USA government is currently spending %17 of GDP on health programs that don’t deliver the results.

    So put an Australian class health system in the USA you would not even need to nick money from weapon making. Just start implementing the key core programs of Australia and kill of the programs that don’t work.

    The first thing to bring you Medicare costs down in the USA would be implement at least a section of the Australian PBS particular the section saying that from drug maker to chemist can only be at max 3 deep. So drug maker to 1 Distributor to Chemist to client should be all a drug has come into contact with. Of course this does not forbid drug maker selling to more than 1 distributor and it does not forbid chemist buying from more than 1 distributor. Just forbids ending up in the mess of drug maker to 1 distributor to 2 distributor to 3 distributor to 4 distributor to chemist to client and this is for medical safety in that 4 distributor chain if product is contaminated how would you locate what one did it.

    So there need to be some regulation. Correct regulation will bring costs down very quickly.

    So the sad reality here is the USA is spending more money that what is required for a good medical system and getting garbage.

    USA people being a coward and fearing socialism and the like without valid grounds results in them not implementing the regulations they need and wasting money hand over fist.

    If you have a government like the USA spending more than 10% of GDP on health and there medical system is not Canada or better system something is really wrong with their government management and regulations.

    If the USA wants to save some money on healthcare implement either the Canadian or Australian systems and everyone in the USA will be better off.

    This is what happens when you don’t make excuses and in fact look at the facts. This happens over and over again. USA people say a huge stack of myths and when you do look at the numbers is mostly that what they have has been done by idiots so they think it cost way more money to-do stuff than it really should.

    So dougman you are already paying enough tax to have a decent medical system. Just your government is wasting the money and failing to deliver it. In fact that a decent system should be 10% or less you are most likely paying too much tax so implementing a decent medical system should allow you to get a tax cut as well. As I typed you are paying more tax than either Canadians or Australians have to pay some of this is that we have medical systems that work not just budget black-hole not delivering results.

  20. oiaohm says:

    http://www.investopedia.com/news/mylans-1100-price-hike-epipen-lawsuit-may-go-nationwide/
    dougman here is the really horible one about the epipen mess. Mylan the maker of epipen has not lifted the price it in fact receives per unit. Mylan has lifted the list price.
    http://www.investopedia.com/news/mylan-launch-cheaper-generic-epipen-myl/
    Yes this new cheaper generic mylan is only expecting to see 50 dollars of the listed price the other 250 dollars eaten up in the system in the middle in the USA.

    Lets look at epipen entering Australia. It comes in from Mylan to a central distributor then to what ever chemist directly. So only 3 parties involved taking their reasonable sized slices seeing only a 100 percent mark-up compared to the price Mylan gets. Yet the USA is 4 to 6 deep with bigger and bigger slices disappearing. How a fully deregulate system can fail badly.

    This kinda tells you why Obarmacare was screwed from the get go. Medical product distribution need to be cost effective and well design so its not being gouged at every step.

    The recent epipen mess in the USA show one of the major differences between the Australian and USA systems. Australia we have a good medical drug distribution system that helps to keep cost down.

  21. dougman wrote, “Americans do not want government meddling in their healthcare.”

    Can’t deal with efficiency, eh?

    “The share of Canada’s overall economy (GDP) going to medicare has stayed in the same 4%-5% range for most of the past 40 years. In fact, in recent years it’s been declining: in 1992, medicare spending was 4.83% of GDP. In 2009, it was down to 4.25%.”

    “the private sector is far more bureaucratic and much less efficient than the public sector when it comes to providing health care.”

    “Canadians are more likely to receive needed care quickly than Americans. Canadians get more physicians visits per capita than Americans, more immunizations, more hospital admissions, and more surgical procedures. They are more likely to have better outcomes, and more likely to be satisfied with the health care they receive.”

    Maybe USAians have been fed a bunch of lies about “socialism”…

  22. oiaohm says:

    https://en.wikipedia.org/wiki/Sales_taxes_in_the_United_States
    By the way, USA doesn’t have a GST. So trying to say Americans have a higher taxes is just ignorant.
    No this is USA idiot who is ignorant who is attempt to come up with another baseless explain to make out the USA has lower taxes when it does not.

    Yes Australia has a GST. USA has per state sales taxes and local surtax that we don’t have. When you average it out with what is covered and not covered with the Australian 10 percent GST compared to what Averaged American pays in sales tax and surtax the result is average USA person pays more than what an Australian pays in GST. Of course there are a few states that are better off but that is not the average.

    Does not matter how you compare Canada and Australia to the USA in tax as long as you take it to the average the USA person is always behind paying more tax.

    https://www.umhs-sk.org/blog/health-care-around-the-world-why-australia-is-1/Caribbean-Medical-Schools

    dougman I am not going to say Obamacare is good because from Australian point of view it badly designed. Thinking by world health organisation rating the Australian Medicare system is rated number 1. So even that our system has a few issues fairly much every other system sux worse.

    http://www.abc.net.au/news/2016-08-25/mylan-epipen-us-prices-hikes-unlikely-to-happen-in-australia/7784700
    $600 a shot you say? You know that there are two pens per box right? And where is the harm in rising prices? Haven’t you ever run a business, probably not.
    Thinking that the same two pack box in Australia sells for 100 AUD without Pharmaceutical Benefits Scheme (PBS) discount from exactly the same maker. Made in the USA shipped to Australia and sold in Australia at a lower price this kinda does not add up right.

    The Australian PBS systems says a drug company is allowed to make a reasonable profit from selling drugs and Australia government will subsidise to make it affordable. Now if a drug company is unreasonable the Australian government reserves the right to temporary void the drug companies patent to put the drug with delivery method out to tender to third party produces. Then the drug company who has the patents can change the third party producers a percentage based on cost of final product up to a 100% of the price of the clone product to them. So fairly much a business making a drug product for medical usage in Australia is restrict to a max of a 100 percent market up or risk having their product made by competitors.

    Price control it really depends on how it implemented. Australian PBS price control is not government set X price and demand produces obey that. Instead government reserves the right to get a third party quote for making the same item then have that party pay you for patents. So if you are using the most cost effective way to product a product and don’t put more than 100 percent mark up on there is no way the Australian government can get a low price supply right thinking the same quality standards will be demanded either way.

    Price ceiling model does not work. Government forced competition model with capped mark-up does work and is quite fair.

    Thing to remember with patents without something like the Australian PBS you don’t have a proper free market in the first place with medical products. Why because X company use X patent to prevent any other company making the product so can gouge the living heck out of people.

  23. dougman says:

    “USA has high tax rates than Australia or Canada yet they don’t have proper government medical.”

    We kinda had “government medical”, but Obamacare is failing as it should. Americans do not want government meddling in their healthcare. Neither of you nimby’s seems to understand that point.

    By the way, USA doesn’t have a GST. So trying to say Americans have a higher taxes is just ignorant.

  24. dougman says:

    “Example: Consider the Epi-pen. It’s a life-saver and rather inexpensive to produce. By promoting a business monopoly the maker now charges $600/shot. ”

    $600 a shot you say? You know that there are two pens per box right? And where is the harm in rising prices? Haven’t you ever run a business, probably not.

    Price controls leads to socialism: https://mises.org/blog/how-price-controls-leads-socialism

    Capitalism TRUMPS Socialism every time. This is why inside every Canadian, they is a American secretly yearning to live free.

  25. oiaohm says:

    https://en.wikipedia.org/wiki/List_of_countries_by_tax_rates
    Robert Pogson no dougman is speaking out his ass as normal. USA has high tax rates than Australia or Canada yet they don’t have proper government medical.

    Why does the USA need to tax so much. Is that USA are pure cowards and invest so much into development weapons instead of taking care of the populations health. In fact the USA has invested more in weapons development than they had so now they have huge massive debts to pay back.

    To make this happen, government would have to increase taxes to every citizen, which can plunge United States into a socialism state where government controls the distribution of goods and services.
    So this is absolutely garbage. If USA had not bankrupted self being cowards the USA might have government medical like Canada and Australia has. Cut percentage of money of weapon development don’t change the tax rate and the USA could have fully paid medical for every citizen.

    Basically socialism was used for the excuse for the cold war and investing more money than the USA had in weapons and USA putting dictators in charge of countries. USA idiots still use socialism as some kind of excuse why they cannot have government medical. Yes the tax rate one is a bold face lie.

    This is normal for a lot of USA people. Lot of claims not one bit of fact.

    https://en.wikipedia.org/wiki/V-2_rocket#Australia
    There is a little thing lost about World War II that explains why we don’t exactly like the USA or UK. Note that that rocket is refereed to as a A4. Those are V-2 rockets built in Australia from stolen German designs to end Japan that never happened. After the war when the USA and UK found out Australia had A4 missiles. A4 missiles have longer range than German V2. USA and UK formed a deal that Australia would not get the German jet engine technology. For Australia to get jet fighter aircraft Australia had to agree to give up the ballistic missile program and still would not be allowed to make jet aircraft to keep us under the thumb. Nothing like being back stabbed by cowards and force to trade security options because of it.

  26. dougman, forgetting that America spans two continents, wrote, “Americans are not interested in paying significantly higher income taxes to have government-provided healthcare, nor do we want government intrusion into private industry.”

    Canadians love it. We live in North America too. There are all kinds of efficiencies in having a single-payer system and if there’s just one payer, it might as well be the government so we can watch over it. Healthcare affects all of us and should not be run as a business. It’s not. It’s a service we can provide for each other by paying taxes. The rich pay more because they can afford it and the poor pay less because they can’t. Then we’re all healthier and living longer. It’s a good thing. I remember the time before Medicare in Canada and people did without healthcare simply because they could not afford it and then they died young. Healthcare, education, military, research, infrastructure are all universally beneficial kinds of things that governments can do well and businesses can’t.

    Example: Consider the Epi-pen. It’s a life-saver and rather inexpensive to produce. By promoting a business monopoly the maker now charges $600/shot. They paid competitors to leave the market. How’s that working for “‘mericans”? In Canada, Medicare makes a list of approved products and procedures and prohibits monopoly, even to the point of insisting on generic products (No-name brands etc.) for medicines. Simple moves like that really reduce the cost of healthcare as does single-payer. We have 11 people on the board for Medicare, just the ministers of Health. In USA, there are dozens of insurance companies charging what the market will bear and until Obamacare were refusing to insure many people and prevented people from changing employer for fear of losing benefits. That’s just silly/wrong. Medicare makes sense.

  27. dougman says:

    Americans are not interested in paying significantly higher income taxes to have government-provided healthcare, nor do we want government intrusion into private industry.

    To make this happen, government would have to increase taxes to every citizen, which can plunge United States into a socialism state where government controls the distribution of goods and services. United States is currently practicing free market system where market forces determine the availability and cost of health care services and not the federal government. The introduction of universal health care will lead to great distortion in the economy. It could be a first step towards government controlling other aspect of the economy.

    The universal health care will reduce the amount of money doctors can make thereby reducing the attraction in the practice. This could create shortage of doctors in the long-run. Those with higher incomes may complain that their money is going to pay for the health care of other citizens when it should be going to pay for their own health care.

    No thanks.

  28. dougman wrote, “You and Robert seem to bash Americans fairly consistently. What, can’t take the heat?”

    USA is a powerful innovative country. However, much of its wealth/prosperity/innovation come from climate. Canada spends much more on heating homes and businesses and does a lot with a shorter growing season. Canada has a lot of glaciated rock whereas USA has huge agricultural zones for corn/beans/fruit/veggies. Both countries have benefited greatly from immigration and USA had slavery which was a benefit for a time but guess where immigrants want to move to, a land of ice and snow or a warm prosperous place wildly proclaimed by Hollywood? So, if USA is spending less time surviving, they have more resources to invest in innovation. Sadly, they still lack universal healthcare and kill each other at alarming rates…

  29. Deaf Spy says:

    Well, You and Robert seem to bash Americans fairly consistently.

    And not only that, Dougman. Robert seems to see Fifi as “technically competent”. Some things defy mind.

  30. oiaohm says:

    Oh yes, the “coward” Americans that fought the Germans and Japanese at the same time and won. Had we not, you would be speaking Japanese now, you’re welcome.
    False USA belief. With the Japan in world war II without USA assistance Australia was 6 months off ending that war. It was the battle of Kokoda Track that decided if Japan got to Australia or not. USA was not involved in that instead the USA left us to our fate.

    The missile platform is still in the top end of Australia where at the end of Word War II if the USA had not dropped nukes on japan by plane 6 month latter Australia would have drop intercontinental ballistic missiles on them with biological payloads it was a 30 missile strike planned to contaminate every single section of Japan. This was the planned revenge for the bombing of Darwin. So the USA saved Japan from Australia by nuking japan.

    Scary part here is Australian intelligence knew the bombing of Darwin was coming same with the bombing of Pearl harbour. The difference was Darwin had evacuation before the bombing yet USA Pearl harbour did not have evacuation even that USA command had been warned.

    BAHAHAAhaha.. One finger death punch, as in like the game on STEAM?
    Cowards punch/king hit is documented real. There is are weaknesses in the human body if you know where they are and hit them you will kill.

    https://en.wikipedia.org/wiki/Sucker_punch USA refers to the style of punch as a sucker punch because it normally performed as a blind side punch. 91 deaths in 4 years due to the punch we call the Cowards punch/King hit in Australia. So the one finger death punch is one of the fake things that stops people thinking about real world killing punches. There are two killing punches one that destroys the brain the other stops the heart and both absolutely real. Both massively fatal even with medical attention. Yes 1 punch killing hit at the door the emergency award is still have over a 90% chance of death. Best medical only gives you a 1/10 chance of living. Way more fatal than a bullet on average.

    So dougman just like normal scared idiot America who demands guns believes the document 1 hit punch kills are fiction when they are absolutely reality. Advantage is both of the one hit kill punches are absolutely silent to pull off and you always have the required weapon to perform them with you.

    America was founded on principles and we fought for our freedom from the crown, Australia was a penal colony of the crown. Huge difference wouldn’t you agree?
    The difference is not as large as the USA idiots want to believe. That skips over a key bit of Australian history that we won our war for freedom quickly.
    https://en.wikipedia.org/wiki/Eureka_Rebellion
    Australians also fought for our freedom it also came clear to the British that something had to be done fairly quickly due to find themselves being cut off from supplies. So in a single battle and cutting supplies from the British the war for Australian independent government was done started being sorted threw the courts by 1986. So we only lost 27 people to have the right to vote in our own government. What happens when you don’t depend on the force of the gun to win freedom. In fact it was put to a vote of the population 1901 if we remained under english rule or not and we could have left without a fight. USA take 4 years to achieve the same kind of thing with a lot more deaths due to depending on the force of the gun. So using gun was 1 year faster really 365 days longer to get it with way lower death toll I think Australia was smarter at it.

    “Bowling for Columbine” mentions that USA person is more likely to die from smog than what they are to die from a gun yet it gets no media coverage. But due to the idiot USA culture pollution control is lower than guns.

    “School shootings are a major problem” name where school shootings have happened in Australia that right they don’t happen.

    Dougman you quoted active shooter there are a lot more accidental due to USA people keeping guns under pillows and the like out of stupid fear that has no base in numbers because they are cowards instead of brave. There is a lot that large percentages of the USA population do that have no base in the crime stats just being scared for no good reason. Those actions lead to weapons being free for children to use in school shootings. USA has a very big coward problem that need to be fixed.

  31. dougman says:

    “Dougman you need to watch “Bowling for Columbine”..

    Why would I watch crappy fake-documentary. Moore’s assertions are patently false, for example, “School shootings are a major problem” when in fact you are three times as likely to die being struck by lightning.

    Americans are 38% more likely to be killed by a lightning strike than by an active shooter. http://www.fbi.gov/news/stories/2014/september/fbi-releases-study-on-active-shooter-incidents/pdfs/a-study-of-active-shooter-incidents-in-the-u.s.-between-2000-and-2013

    “The reality is the Americans are land of the cowards.”

    Oh yes, the “coward” Americans that fought the Germans and Japanese at the same time and won. Had we not, you would be speaking Japanese now, you’re welcome.

    “The reality here why we don’t need guns.”

    Yep, you certainly do not, you are too incompetent.

    “Most Australians know how to perform what is called now the cowards punch or was called the king hit. Yes 1 punch death move. If you were not brave you would not walk around in public in Australia knowing that every single person with a fist is armed with the skill to kill you.”

    BAHAHAAhaha.. One finger death punch, as in like the game on STEAM? http://store.steampowered.com/app/264200/ LMAO..I know this may come as a shock to you, but things that happen in your mothers basement, do NOT reflect the REAL world.

    “From the time of the pilgrims to now in the USA the population of the USA has not over come being scared idiots who take reprisals and demanding weapons instead of balanced and correctly researched response as a civilised country should do. ”

    Ham-Dong, the historian! Oh, do tell what incidents you are referring too.

    Well, in response to you illogical statement, America was founded on principles and we fought for our freedom from the crown, Australia was a penal colony of the crown. Huge difference wouldn’t you agree?

  32. oiaohm says:

    Dougman you need to watch “Bowling for Columbine” made by an American.
    Well, You and Robert seem to bash Americans fairly consistently. What, can’t take the heat? And you call yourself an Australian; you’re nothing but a tosser!!
    Cannot speak for Canadians. But I can speak for Australians. Every year being called a tosser by a person from a different state of Australia is perfectly normal. We get over it and laugh about it latter on. Because we truly live up to the idea of the land of the brave. The reality is the Americans are land of the cowards.

    The reality here why we don’t need guns. Most Australians know how to perform what is called now the cowards punch or was called the king hit. Yes 1 punch death move. If you were not brave you would not walk around in public in Australia knowing that every single person with a fist is armed with the skill to kill you.

    From the time of the pilgrims to now in the USA the population of the USA has not over come being scared idiots who take reprisals and demanding weapons instead of balanced and correctly researched response as a civilised country should do. Sorry dougman I can take the heat but can you take the heat about the real defects of your country.

    You don’t get it what we do is for your own good as your countries culture need to grow up and stop being a stupid spoiled brat.

  33. dougman says:

    “dougman don’t comment on cultures you don’t understand. ”

    Well, You and Robert seem to bash Americans fairly consistently. What, can’t take the heat? And you call yourself an Australian; you’re nothing but a tosser!!

  34. oiaohm says:

    dougman don’t comment on cultures you don’t understand. White is lighting/sky gods. Black is earth gods. Yes it the earth gods that are shaped like a gollywog. So yes Australian Aboriginals pain themselves either black or white and other colours depending on what part they get in the dance. Rainbow serpent one of the forces of creation is interesting.

    There are other variations. “Bawoo Stories How crows became black” Would be a good one for you to read dougman and wake up there is a dance to that one. So yes you start the dance white but by end of dance you have applied charcoal based colour to end the dance black head to toe. There are also dances that are the reverse of that so starting black coloured and ending white colour guess what you had a picture of.

    Now as a white person get a lighting/sky god position in a dance is a lot lighter on body paint requirement like a dark native Australian Aboriginal getting a earth god role. Yes Australian Aboriginal native skin colour is highly variable to as light as a Caucasian that has been lightly sun burnt to really dark and that is not due to interbreeding just natural full blood colour variation. So to perform dances correctly for a long time they have applied their versions of make up.

    As Dougman said
    Typical dweeb. Lets change the subject now, as I painted myself into a corner yet again.
    Funny bring up this aboriginal stuff as dougman said himself is that attempting to change the topic like this is a Typical dweeb.

    So the last 4 posts by dougman are that of a idiot dweeb by his own standards.

    So just like with the fact ltsp meetings not being run any-more with one bit of incorrect information dougman thinks he knows everything when he knows nothing.

  35. dougman says:

    Ham-Dong, always having a excuse for something.

    For example, I thought it was the scientific method you use? Or maybe its dyslexia? Or English is not your original language, or some other lame excuse.

    Go pound sand with your aboriginal tribesmen!

    By the way, is it not funny that the natives paint themselves white, but ahem, you stated black. Perhaps you are color blind to boot?

  36. dougman says:

    You know, this is a super big maybe. I have mentioned it before but the process is a pain in ass.

  37. dougman says:

    Sorry it attack my IQ again because you were pulled up for being completely wrong again.

  38. dougman says:

    “This is the problem with the 24 hours delay before clean proof reading even sometimes where I spot an error sooner I see it in the wrong place. It was a typo and then incorrect fix. You have to admit it happens a lot less often these days. And that is because I have been posting stuff. So I have a reason to do posts.”

    Since I suffer from dyslexia I have to understand the diagnostic methods used for it and the statistics about it.

  39. oiaohm says:

    DrLoser thank you for picking that one up.
    Dougman you have painted yourself in a corner you accused ltsp of being inactive that was not the case.
    This is the problem with the 24 hours delay before clean proof reading even sometimes where I spot an error sooner I see it in the wrong place. It was a typo and then incorrect fix. You have to admit it happens a lot less often these days. And that is because I have been posting stuff. So I have a reason to do posts.

  40. DrLoser says:

    Dougman you have painted yourself in a corner you accused ltsp of not being in active that was not the case.

    opps typo.

    Classic Fifi, from beginning to end.

  41. oiaohm says:

    Robert Pogson even subtracting just the hard drive from every individual machine and using a central file storage like cdboot, drbl and ltsp (in fat-thick- client mode) is a reasonable saving in maintenance. Ltsp does these days both thin and fat(thick). This is where comparing ltsp to windows terminal server does not add up any more ltsp has a feature windows terminal server lacks.

    Biggest issue with the traditional desktop/individual OS installs in large numbers is tracking down the odd machine that
    1 was not turned on so has not received updates
    2 had some other issue with a update so now not install any other updates
    So machine does not end up updated as machine should be and has to be hunted down this costs time. This issue happen if the traditional desktop OS is Linux, Windows, OS X…. even stuff like chromebooks you have a case that students are running different versions of the OS because some of them have not updated.

    One file-system needs maintenance instead of ~25.
    Minimum number of file-systems and operating systems to maintain is the big one.

    Thick clients are normally slightly heavier on ram since they normally cannot depend on having any swap space.

    The clients exhibit the performance of the server which was often newer/faster/bigger by an order of magnitude so everyone had great performance for most tasks.
    The key word is most. The reason for the hybrids between thick and thin is for the tasks that don’t perform better in thin-client mode yet still have reduced maintenance issues and proper central backups by going thick client mode.

    Remember organisations are willing to pay more to get ccboot to run windows as single image to get out of the maintenance hell. So there are two savings here. Pure thin can be cheaper in up front dollars and thick clients are lighter maintenance. What fairly much leaves no space for the traditional desktop computer other than cases like windows deployment where you have to pay extra for the thick client mode.

    Portable devices you kind have no choice they have to run self contained but that does hurt you in maintenance.

    This is where some people make a big mistake in TCO. Every extra individual OS install costs you extra in maintenance. Even when using cloned images.

    If you have 202 machines and you end up maintain 3 OS images being two OS server installs and 1 OS image for the 200 clients you are laugh compared to having to take care of 202 OS images due to individual installing and the variation that will happen.

  42. oiaohm wrote, “LTSP has it usages particular with central managed machines to let students code on that if they stuff it up its a simple reboot to fix most of the time”.

    When I was teaching, the beauties of LTSP were:

    • One installation sets up the whole lab if the clients boot from the server.
    • One file-system needs maintenance instead of ~25.
    • The clients exhibit the performance of the server which was often newer/faster/bigger by an order of magnitude so everyone had great performance for most tasks.
    • Total cost of hardware even if purchased new could be a fraction of the cost a typical lab of thick clients, way less power, RAM and storage needed to run a lab. The difference was huge if you bought/built one new server to serve a bunch of old/slow discarded or existing clients.
  43. oiaohm says:

    dougman that is the thing you missed about ltsp. Is the minutes posted is short version.
    http://irclogs.ltsp.org/meetings/ltsp-meeting/2013/ltsp-meeting.2013-04-15-18.31.log.txt
    You only read the wiki page right. In the full IRC log of the 2013-04-15 you notice Debian EDU project and Edubuntu taking over general operations so no future ltsp only meetings scheduled. Instead there are joint meetings between both projects covering all items relegated to education this include ltsp. It is fairly much don’t run too many formal meetings when you don’t have to. Then in 2014 education only meetings went by by.

    In fact the same thing could happen to reactos if it was taken in by another project.

    I did say that distributions had taken over maintain-ship ltsp is in fact in the last ltsp only formal meeting. Since 2013 ltsp has only had Ltsp only informal meetings. Exactly what use is talking about altering ltsp if those alterations would not work with education software where ltsp is most commonly used? Does it not make more sense todo it in the formal meeting over education software. It did not make a lick of sense for ltsp to keep on running it own isolated formal meeting.

    This is a catch just because a project is not running it own formal meetings and more does not mean that something has not happened that makes for that project running own formal meetings completely pointless even that the project is viable.

    Dougman you never explain why ltsp had to run formal meetings. In fact you never considered if it was possible that something that happened that made running own formal meetings pointless.

    Mind you if you want example of something that has not run formal meetings for a down right long time. Last formal bind9 dns formal meeting was 2003. Yes the most common deployed DNS server has operated on a pure informal meeting method for now 13+ years.

    Linux kernel has not seen a proper formal meeting in 12+ years.

    Counting how often a project runs formal meetings shows absolutely nothing about how viable project is. There are quite a few viable projects that have never once run a single formal meeting. Only a complete idiot thinks how often formal meetings happen tell you anything about the viability of a project.
    Last

  44. dougman says:

    Ignoring all your other senseless ramblings, I did notice something of interest.

    https://www.reactos.org/project-news/september-2016-meeting-minutes

    Not only do they meet on IRC, they actually post a listing of minutes! I can say that ReactOS seems like a far more viable project then LTSP, but not that it’s of any use to anyone.

    Awaiting your spam post, explaining why ReactOS is of use and other silly nonsense.

  45. oiaohm says:

    So, yet again, you did not contribute anything.
    The reactos build environment construction script is based off one of mine.

    dougman I had done a lot of pure cross building mingw on linux without using any binary parts. The method in is ROSBE is mine. So sorry Reactos so something I have contributed to under this handle. Sorry big mistake presuming I have not contributed anything. I have made many contributes to many different projects. I very rarely mention any of them. If you had bothered looking you would have found out me contributing something is not isolated event. And I normally do not talk about it. This means I don’t not fit what you are in fact calling me. So I downplay my accomplishments most of the time.

  46. dougman says:

    By the way, our resident Doctor hit the nail on the head, naming you FiFi.

    Definition: fifi
    Prevalent with inmates, a hands-free masturbation device made up of a dirty old sock, vaseline (or toothpaste, if not available) a toilet paper tube (if available), wedged firmly between a matress.

  47. dougman says:

    “Look at a debate in reactos”

    Oh dear God!…. you are not trying to resurrect that mess again are you?

    “leads to me looking into something.”

    So, yet again, you did not contribute anything.

  48. oiaohm says:

    dougman for fair disclosure I was required to state how I know the IRC handles when reading the log. Going to what feature I added to ltsp would be going off topic. So you have said I should not add other topics and this is exactly what you are asking by demanding I show what I added.

    You are just after a reason to personally attack me because you don’t want to admit you said a stack of lies.

    Furthermore, you said “now is part of Ltsp. So its something hidden in irc log history” of course it is, hidden from view means you did not contribute anything, you’re just making up crap to suit your delusional narrative.
    This is because you are in fact the delusional one dougman who thinks a person has to show everything they did. Look at a debate in reactos forums you would find a debate that leads to me looking into something. And that is under this handle then cross reference that with the IRC ltsp you will find my other handle and my bad writing style. So it not hidden form view just hard to find and it the way I want it.

    They may exaggerate accomplishments in personal or professional areas in order to make their lives seem impressive.
    Sorry in fact I do the other I normal down play what I have done.

    If I was wanting to make myself seam impressive I would have stated what the feature that was added because of me. Because it quite a good one. Do a care about the fame I could get for the feature. Not at all. Key thing was having the feature.

    Basically this is just personally attacking me for personally attacking me because that is better than admitting you are incompetent dougman. The reality with what you accused ltsp of means you cannot background check properly to save yourself.

  49. dougman says:

    “So yes I did hand over source code. ”

    MORE BS. What programming language and editor did you use? See no one believes you and your BS, as you say one thing then say another.

    Example: ““So I demoed a feature in a hacked up form.” no where in that statement does it say that you uploaded/edited/created anything, then you go one to state that, “That feature was rewritten by one of the maintainer” which doesn’t imply that you wrote anything, reinforcing what I stated to you.

    Furthermore, you said “now is part of Ltsp. So its something hidden in irc log history” of course it is, hidden from view means you did not contribute anything, you’re just making up crap to suit your delusional narrative.

    I understand though, truly I do, when people have a low self esteem of themselves, it commonly known that they tend to make things up. Pathological liars, which you seem to fit the bill to a tee, lie to make themselves look more important than they truly are. They may exaggerate accomplishments in personal or professional areas in order to make their lives seem impressive.

    Honestly, you may be able to pull this off with a naive female, if you weren’t so ugly, but trying to push your BS story on people that know is ludicrous; perhaps you need a check-up from the neck-up. Know what I am saying?

  50. oiaohm says:

    Dougman you have painted yourself in a corner you accused ltsp of not being in active that was not the case.

    opps typo.

  51. oiaohm says:

    http://mrpogson.com/2016/10/06/amazement/#comment-354268
    Typical dweeb. Lets change the subject now, as I painted myself into a corner yet again.
    No I answers you post.

    Dougman you have painted yourself in a corner you accused ltsp of being in active that was not the case. I know this because I had worked with ltsp quite a bit. I did not change the topic you did. So by your own standard you are a typical dweeb who every time you lose you add another topic to attempt win. I have not added a single new subject here at all. So I have never changed the subject. Only person changing the subject is you Dougman.

  52. dougman says:

    Every time you post on here, your mind goes splat.

  53. dougman says:

    Typical dweeb. Lets change the subject now, as I painted myself into a corner yet again.

    Avoiding my questions: http://mrpogson.com/2016/10/06/amazement/#comment-354263

  54. oiaohm says:

    By the way the failure with graphical Ubuntu Linux applications going splat under WSL is an related reason why a lot of Android applications went splat under “Project Astoria”.

    Android applications it was Opengl ES and Ubuntu Linux graphical applications it is mostly Opengl. Its how to get a GPU data from inside WSL/”Project Astoria” to host OS Windows and back again that Microsoft Developers never solved. I will say that is not a easy problem to solve. So yes WSL still suffers from the same faults that caused “Project Astoria” to be scrapped. WSL allowed the “Project Astoria” developers to save some face for the expenditure the failing project was costing.

  55. oiaohm says:

    I doubt they could get away with running Android apps on top of Windows, but working more with Linux seems to be their goal long-term.
    dougman WSL comes out of Microsoft failed attempt to run Android applications on Windows phone called “Project Astoria”.
    http://www.windowscentral.com/microsoft-officially-cancels-project-astoria-bridge-porting-android-apps-windows-10-mobile

    Not they Microsoft could not get away with running Android apps on top of Windows they in fact did. The issue was number of applications that would bite the dust.

    So it was a worthless contribution.
    Worth of contribution was not the point here. To demo something when all you have is irc you have to hand over how you did it so they can do the same thing. So yes I had to create the scripts for it and gave them scripts for it. So yes I did hand over source code. Not all source code handed over ends up in mainline and that is the reality. Not worthless because the feature I had implemented is now implemented better.

    Wait, you just stated you “demo’d” something, now you are stating that they “rewrote” your demo?
    This is a failure to understand what demo can mandate. This form of demo where you have handed over source code does happen to get rewritten fully at times and if it does you don’t get a commit entry. Now if it was 95 percent rewritten and 1 line of mine remained I would have got a commit entry. So not everyone contributing a new feature to a project gets a commit entry.

    Dude, get your facts straight. You lie so much, that you are starting to believe your own lies.
    Dude get you mind around it that words depending on where they are used mean different things. Demoed on IRC means a little more than just showing pictures/video to someone. Basically its your lack of understanding of diversity english is leading to you accusing me incorrectly again of lieing.

  56. oiaohm says:

    Like it not dougman and Drloser all the responses to this topic there is not a single lie by me. If anyone has been lieing is been dougman and Deaf Spy.

    Remember dougman has attempted to make out that 2 project so far are not alive when they are because they are not functioning the way he expects. Deaf Spy suggested a impossible to work properly solution due to technology defects.

    I defy anybody to show me a use case. It was a bright idea that had its day, but that day is past.
    DrLoser I already did. http://pinet.org.uk/ Starts 22 Sep 2013 as Raspi-LTSP. With new deployments though 2014, 2015 and 2016 not recorded Ltsp usage map. Yes the core to pinet functionality is Ltsp. There are another modern use cases of Ltsp if you look around also not reported on the Ltsp map of usage locations. So this is you posting without reading what I have already posted.

    Useful technology that is low cost has habits of not going way just changing it usage form.

    So you asking for something already done and dusted and it was something

    Dougman attacked pinet because done differently. If you had watched a particular 2015 python conference video you would see that the recommend item to use for documentation is not wiki or forum but git these days because someone surveyed open source projects for what ones had the best documentation. Turns out running a wiki or forum in most cases decreases the quality of the projects documentation due to lack of review. So some of pinet differences is that it using more modern methods of open source project management. Little bit of a shock to system to those who are not use the modern design for running open source projects.

    LTSP has it usages particular with central managed machines to let students code on that if they stuff it up its a simple reboot to fix most of the time with the PI devices at worse swap the SD card and rewrite the SD card because a PI does not have built in PXE. There is a downside to have the OS of a computer/thin-client stored inside the device instead of on a network server when you are letting users do random things. DRBL is the other open source solution for OS central server stored but then run on client end.

    There is a solution for Windows http://www.ccboot.com/ but that requires paying a monthly fee(that at best is half what as windows 10 enterprise monthly subscription costs) and at times groups getting into windows licensing trouble due to client machines not having prior copies of windows. If Microsoft ever released something like ccboot, DRBL or LTSP as part of windows the desktop computer in schools would be absolutely dead so fast it not funny replaced by the thick/fat client. The desktop computer has had it day in large deployments.

  57. dougman says:

    Speaking of “Amazement”, in that Robert has grown tired and weary of writing on his blog.

    Here is a decent piece to read: https://www.slightfuture.com/how-to/x-on-wsl

    I am not sure how to take that after reading it, does it mean M$ wants Linux to run atop Windows, or will they be merging the two OS’s at some later time-frame? I doubt they could get away with running Android apps on top of Windows, but working more with Linux seems to be their goal long-term.

  58. dougman says:

    “Did what I contribute end up in the form I contributed main-line the answer is no so there is no commit log entry.”

    So it was a worthless contribution.

    “So I demoed a feature in a hacked up form.”

    LOL..you demo’d something and previously stated you made a contribution? LMAO…you’re so lame.

    “That feature was rewritten by one of the maintainers and now is part of Ltsp.”

    Wait, you just stated you “demo’d” something, now you are stating that they “rewrote” your demo? Dude, get your facts straight. You lie so much, that you are starting to believe your own lies.

  59. oiaohm says:

    I did not find a oiaohm or peter, so obviously you a lying fraud! Wow…who knew?
    Is oiaohm my only handle the answer is no not.

    Did what I contribute end up in the form I contributed main-line the answer is no so there is no commit log entry. So I demoed a feature in a hacked up form. That feature was rewritten by one of the maintainers and now is part of Ltsp. So its something hidden in irc log history why I knew who was who in that channel. Of course I am not going to give either of you idiots another handle of mine.

    I mainly mentioned it DrLoser because I was sure as normal dougman would fail to check all the correct locations for contributes. So checking if someone has contributed to a open source project is quite a hard process when you remember that at times what is contributed is altered before it mainlined.

    DrLoser have you not forgot the rule call me fifi and ask for a cite I will never provide the cite. Sorry dougman blame DrLoser for me now not tell you what I contributed ltsp.

    http://www.ltsp.org/stories/viewstory/?story_id=8&secret=ed1d48
    DrLoser you need to look deeper at that greek schools one. Implement 2007 reported as existing 2013 so 6 years latter. Reason for the report is part of the IMF forced software audit and reporting started in 2010. So without IMF intervention that 180 greek schools locations using Ltsp we might not know about even today. Yes those 180 locations were deployed before the Greek debt criss. So dougman statement that the massive greek usage is because of the debt criss is incorrect. Greek reporting of usage is because of the debt criss. Brings the question how many other locations who are using Ltsp are not reporting it.

  60. dougman says:

    “dougman I know who is who with LTSP because I did contribute something about 6 years to LTSP”

    Did you write a text file?

    https://www.openhub.net/p/ltsp/commits

    I did not find a oiaohm or peter, so obviously you a lying fraud! Wow…who knew?

  61. DrLoser says:

    dougman I know who is who with LTSP because I did contribute something about 6 years to LTSP.

    Did you indeed, Fifi?

    I notice you cite various other things directly below, eg Ubuntu. For some inexplicable reason, you chose not to cite your “contribution” to LTSP.

    Pray remedy this sad lack of honesty.

  62. DrLoser says:

    Wouldn’t it be ironic if these few LTSP deployments, are actually just RDP gateways to Windows servers.

    That’s a very good question indeed, Douglas.

    One further very good question is — why would anybody in 2016 even care about LTSP?

    I defy anybody to show me a use case. It was a bright idea that had its day, but that day is past. To quote from the LTSP website:

    The Greek primary and secondary schools that are using Linux Terminal Servers have been added to the LTSP world map! With over 180 different locations and LTSP setups, this is without a doubt one of the largest LTSP implementations in the world.

    A whole 180 locations, in Greek primary and secondary schools.

    Warm gl0w here …

    Oh, and the usual completely specious and irrelevant Ham-Bone “cites.”

  63. oiaohm says:

    dougman I know who is who with LTSP because I did contribute something about 6 years to LTSP. So being your normal unresearched idiot with an attempted insult by wild guessing.

    Wouldn’t it be ironic if these few LTSP deployments, are actually just RDP gateways to Windows servers.
    That is rare and it shows you not knowing topic.
    https://help.ubuntu.com/community/UbuntuLTSP/VMWareViewVDI
    Why VDI not windows terminal server. Windows terminal server solutions opengl and direct x acceleration does not work without adding on third party add-on that has a habit of snapping with windows updates.

    So a few LTSP deployments are gateways for VDI solutions that in fact have acceleration working with Windows. It is helpful to be able to update you VDI software once and then just reboot all the clients instead of having to flash all the clients and reboot them and pray you have not bricked any.

    Terminal services have evolved. More and more LTSP deployments are what are know as thin/thick hybrids clients or fat clients. Pinet is example of a fat client where application do in fact run on the client directly just the software is not stored on the client other than enough so it clients can get software from the network.

    512 Meg ram Raspberry PI is good enough for most of the VDI and RDP client applications. http://rpitc.blogspot.com.au/2016/07/rpitc-v3-101-release.html

    So yes just because somewhere is using LTSP to manage thin clients does not mean users ever see Linux. The reality it does not matter if it XEN or ESXI providing the VDI both of those solutions a core part is a Linux kernel. So windows or linux desktop VDI to users the server will normally contain something Linux and for easy thin-client updating from a central server LTSP can be deployed.

    Thin-client network can be anything from fairly simple to maintain to a complete pain in but particularly if a server side updates and you have to flash each of the clients with new client software so they work stable. LTSP and DRBL versions of a thick client/fat client is more friendly than most flashed in thin-clients due to a central servers hosting what the clients are running making it update once to update everything and not risking having flash failures.

    So LTSP is a fairly useful tool if you are going the wired thin-client route. LTSP is also a useful tool using it fat mode to allow people to modify system configurations with out effecting other people who log onto the machine.

  64. dougman says:

    “vagrantc is the debian package maintainer. alkisg is ubuntu package maintainer.”

    But no contributions by good ol’ Ham-Bone.

    Wouldn’t it be ironic if these few LTSP deployments, are actually just RDP gateways to Windows servers.

  65. oiaohm says:

    http://irclogs.ltsp.org/?d=2016-03-25
    dougman the day of the last version change.
    http://bazaar.launchpad.net/~ltsp-upstream/ltsp/ltsp-trunk/revision/2731
    Informal meeting alkisg and vagrantc for release testing.
    vagrantc is the debian package maintainer. alkisg is ubuntu package maintainer.

    There should be 36 meetings or so for you to list out. There is closer to 600 informal meetings over that 3 year time frame. So lots and lots of informal meetings. So no way I am going to list them all. I even gave you a clue to what date to look one up when I pointed to the release file last change. Of course normally before release there is either a formal or a informal meeting for people to yell if it busted or not. Finding one proves you were making a mistake. Complaining about them not running formal meetings is valid. Saying no meetings is wrong.

  66. dougman says:

    “Dougman the different patches tell you the date and time of the meetings.”

    Oh?? Okay, please list out all the meetings since 2013 to 2016 then. Be sure to include all revelation patch dates, that coincide with such meetings.

    There should be 36 meetings or so for you to list out.

  67. oiaohm says:

    Trying to insinuate that IRC logs equates to meetings? Where are the meeting minutes, notes, presentations, graphs?? You know, the common things that show up in a meeting.
    Dougman the different patches tell you the date and time of the meetings. The irc logs kept the minutes of those meetings. Any who knows meeting process knows that notes, presentation and graphs of meetings does not have to be archived. So this is just you being an idiot asking for stuff that does not have to be kept. A formal meeting has a chair person/op in irc case controlling who can speak when. Informal you don’t need assigned chair person or clearly define meeting times just that the meeting happened and minutes were recorded somehow. Yes it always important when saying no meeting held to state if it is informal or formal or both. Ltsp case formal meetings have not been held is true but informal meetings have been held.

    You know what institutions like schools and SMEs do these days to allow for redundancy, Robert? They use WiFi.
    https://www.newcastle.edu.au/current-students/campus-environment/information-technology/computers/wireless-or-wired
    DrLoser the idiot. School do use Wifi yes but they also do deploy wired. Lot of wired. Why when students all attempt to submit assignments on the same day you can fairly much say the wifi will be stuffed.
    https://www.bufferbloat.net/projects/make-wifi-fast/wiki/
    All the issues talked about with this project is why going pure wifi falls in on your head. With the redesigns of wifi radio usage it might be practical to deploy wifi without wired lines in about 4 years time. Not today the issues are road block.

    OK then, show me a ChromeBook that can’t deal with WiFi.
    Its not the chromebook that has the big issue itself but it causes some. It the wifi routers method of sending traffic and hearing traffic from other machines due to radio noise. Lot of Chromebooks transmit at full strength with wifi all the time. This comes a nightmare for density.

    Lot of small SME use wifi but you are talking like 3 channel wifi routers and less than 12 machines so it works. Schools and universities it number of devices involved means you hit all the wifi defects.

    Call me naïve, but “is no longer needed” is a fair complement of “defunct.”
    DrLoser something that is no longer needed as independant does not equal defunct. No longer need as independent download is ltsp case. The linux distributions themselves now take part in directly integrating ltsp and providing their own support pages.

    https://www.healthcarelink.com.au/index.php?post_permalink=1&post_id=practice-nurse/11405
    So to finalize. Doctors don’t go picking fruit, unless they are a loser. LOL…..
    DrLoser of course backing something like that. Dougman is a idiot people as stupid as it sound from all professions go picking fruit as a holiday. Not that I understand that logic but they do. The one thing about fruit picking its normally not stressful.

    The worst lab had network switches stacked up in a wooden box in a corner and the cables ran over the floor in conduits constantly tripping students as some idiot arranged the table to cross the conduits.
    Yes Robert I have seen these kinds of mess. One of them was they run out of budget for a false floor then designed to go forwards with cables in conduits place as if there was a false floor. Stuff like this makes duct tape cabling look good. Duct tape cabling is good when you have the case that all building modifications have to go through a long approval process as well. Duct tape is classed as temporary so can be done instantly. Yes it would have been safer duct taped to the floor than in conduits on floor. Cable lose on floor and conduits running across open floor are equal tripping hazards. Duct tape cable to floor the cable does end up failed due to being walked on but never normally trips anyone.

    Duct tape cabling is great for cabling an area then testing if your design is right before spending on conduits as well.

    Stapling to the woodwork and not being under foot does under mine the network transfer speeds a bit but at least its like duct taping being fairly safe. Most of the stuff about cat 5/6 with kinking, crimping, bending sharply to so you don’t snap wires in deployment and get full network speeds but cat 5/6 has quite a bit of tolerance designed in. Cat 3 that cat 5 and 6 is based off was designed with the theory of deploying it with a staple gun so the basic wall thickness of the cat 5 and 6 cable is in fact design to take as long as done right. This is why you find cases where stapled up Cat5 and cat6 are working ok. The trick is you start stapling at one end of the cat5/cat6 and work to the other end and maintaining at least 35cm between staples and not wanting to send signal more than 20 metres(yes 1/5 the distance of doing it right with plastic holding clips). Don’t use metal clipping is about not causing signal interference at 100Mbps+. So the limitations mean you should not. Desperate is another matter sometimes should not is done and works if person knows how to correctly do the should not.

    If a person cannot duct tape cable into safe locations don’t pay them to fit conduits.

  68. DrLoser wrote, “doubling up the amount of Cat-V (or pick other) Ethernet cabling is a fairly expensive and rather unlikely expedient for yer average school.”

    I think two cables and Ethernet jacks were standard in the Arctic. At first, the idea was that one or two PCs would be enough for any classroom, then along came Computers for Schools and we had several classrooms with 2-5 PCs. I can’t remember what switches we used as there was no particular budget but I suspect it was obsolete switches perhaps 10mbits/s left hanging around. Might even have been hubs… A few years later desktop switches became cheap, plentiful and faster and any teacher could just throw one in and fan out as long as the DHCP server and router were OK with the new machines. In most schools they were. It was pretty easy to slip new/old machines onto the networks. CFS just required a principal’s signature and freight in remote places. In a few places, teachers agreed to bring them in as luggage so no freight was charged. In other places local businesses agreed to donate freight. e.g. retail stores could bring in 20PCs along with tons of food either as sea or air-freight. Servers were more tricky. Some schools had ways to insert a server into the budget just on general principles of upgrading but in other cases a nice new PC would serve and no one at Head Office needed to know it was to be used to run an LTSP lab or file/database/print/web servery. We often used a server as a web-cache or local web server to offset our terribly slow connections.

  69. Deaf Spy wrote, “With all this in mind, the total length of cables will easily exceed one mile.”

    Not good at maths, eh? A typical classroom is about 32X32 feet. It can hold a strip of clients around three sides and one or two rows down the middle. There is an arithmetic sum for each of the branches. You might need a 40 foot cable to reach the farthest clients, a 36 foot cable to reach the next closer, etc. The sum is N/2(first + last) ~ 16/2(64+16) + 8/2(24+8) ~ 700 feet, just one roll of cable. Now it’s a different matter if the IT-guy insists the switch be down the hall in a server room. For a lab that is self-contained except for Internet connection, that’s not necessary. I’ve seen both layouts and usually the distributed switches had way more PCs and far fewer IT-guys. Indeed, a lot of schools I saw had the server room built into the lab either just in a corner or in a proper closet. That works. In my biggest network we needed two server rooms and the various classrooms had a local switch with enough ports to do the job. I think most classrooms had 8-port gigabit/s switches and the lab used multi-seat X to expand the network to several seats per PC.

    I’ve seen some horrors. The worst lab had network switches stacked up in a wooden box in a corner and the cables ran over the floor in conduits constantly tripping students as some idiot arranged the table to cross the conduits. I rotated the lab to have the conduits pass under the rows of tables but it was still pretty silly. I converted that lab to LTSP and used all the discarded machines in the school as thin clients, raising the complement of working machines from 14 on a good day to 24 24×7. That was a bit tricky as some machines were more than a decade old and needed extreme measures to get monitors to work but it did work. A further complication was keeping TOOS as a dual-boot. I don’t think that was ever used again but it was an option. GNU/Linux worked for them.

  70. oiaohm wrote, “Yes roll of duct tape and stick the cable to the wall and ceiling.”

    The first lab I had as a computer teacher had an even more wonderful solution, stapling to the woodwork. I kid you not. It looked horrible and dangerous but the damned thing worked. That was set up by the head of medium-sized school division in the North. The techs were embarrassed when I asked about it. Kinking, crimping, bending sharply are all discouraged for twisted-pair stuff but that lab had it all and everything was crimped into place with staples into wood.

  71. DrLoser wrote, “You can do this with ChromeBooks and a proper server.
     
    You can’t with a Cello, as far as I can see.”

    A Cello is a proper server. I don’t see any reason a ChromeBook could not work with a Cello. Just install whatever web-application you want on the Cello along with databases, networking and storage. It’s all good. A Cello would also make a good gateway to the Internet so other web-applications elsewhere could be accessed. It’s not a problem except for obtaining one so far…

  72. DrLoser wrote, “Unsubstantiated useless irrelevant gibberish, Fifi.”

    One of the best places to mount a WiFi access point is in a window to minimise material between transceivers. Duct tape and masking tape do a fair job. My best choice is using duct tape to hold the AP in position temporarily while the silicone sealant hardens…

  73. DrLoser wrote, ” They use WiFi.
     
    Then again, you wouldn’t know, would you?”

    On the contrary, I did set up WiFi in one school just to take care of two broken cables. The alternative was me crawling around in the crawlspace… That’s fine for a couple of clients. It’s much more expensive and troublesome for a whole school. BTW, at that school, someone stole one of the WiFi access points. That doesn’t happen with cables built into the walls.

  74. dougman says:

    “I am not a Loser — I just play one on the Internet — and as such, I can offer evidence in your favour.”

    LOL…mea culpa. No offense was suggest towards you. You just reinforced my argument to HamBone though, with his brain-fart notion that jobless doctors pick fruit.

  75. DrLoser says:

    One other thing to back up Deaf Spy, btw.

    Now, I am a programmer, a designer, an architect, call it what you will. I know nothing very much about IT infrastructure … just about enough to pour derision on Pogson’s feeble efforts.

    But even I know that redundancy is the name of the game.

    And even I know that doubling up the amount of Cat-V (or pick other) Ethernet cabling is a fairly expensive and rather unlikely expedient for yer average school.

    You know what institutions like schools and SMEs do these days to allow for redundancy, Robert? They use WiFi.

    Then again, you wouldn’t know, would you? Because you’re basically a technological dinosaur.

  76. DrLoser says:

    Deaf Spy I have been to a lot of schools I can tell you duct tape is popular.

    Unsubstantiated useless irrelevant gibberish, Fifi.

    Nice to see you’re back on form. No linkies? No estimate as to “a lot of?” No particular reason to visit those schools, out in the back of beyond?

    You do realise that your statement marks you out as a pathetic little dweeb, I assume?

  77. DrLoser says:

    So to finalize. Doctors don’t go picking fruit, unless they are a loser. LOL…..

    Now, now, Dougie. You are too much of a gentleman to come out with unsubstantiated personal mockery like that.

    Let me help you out here by substantiating said mockery.

    I am not a Loser — I just play one on the Internet — and as such, I can offer evidence in your favour.

    I don’t go picking fruit, either.

  78. dougman says:

    “So shut up about stuff you know not, dougman.”

    Lets talk about this anger of yours Robert. I think the best course of action would be to participate in a Anger Management Group Course.

    I know the world of IT is not going in the direction you like, but gosh darnit people don’t see things your way.

  79. dougman says:

    “The meeting was not in person but was in freenode #ltsp-meeting in 2013 The main freenode #ltsp is fully log. http://irclogs.ltsp.org/

    LOL…hammy you are too much!

    Trying to insinuate that IRC logs equates to meetings? Where are the meeting minutes, notes, presentations, graphs?? You know, the common things that show up in a meeting.

    I am unsure about you, but I have sat in far too many meetings, and by far there is always some material handed-out or subject to be discussed. At the conclusion of said meeting, an after-action report is submitted for those not in attendance.

    So to finalize. Doctors don’t go picking fruit, unless they are a loser. LOL…..

  80. DrLoser says:

    Deaf Spy wrote, “defunct project LTSP”.

    Nonsense. LTSP is part of many distros now and is no longer needed.

    Perhaps you might care to consult your trusty online Websters dictionary, anno 1828, Robert.

    Call me naïve, but “is no longer needed” is a fair complement of “defunct.”

    Thus, no nonsense … except in your ill-considered response.

  81. DrLoser says:

    That’s not true and not what I wrote. Some schools, particularly older ones, may not have any cabling except TV/video and may need WiFi. Newer schools should have built-in cabling to Cat-6.

    “Should have.” And indeed “may very well have.” But you don’t limit your choice of school server based upon anything as pathetic as this, Robert — not in 2016. Take, for example, Douglas’ choice of a low-end ChromeBook as a (slightly overpowered) thin client. School districts seem to be quite keen on ChromeBooks. Indeed, you, Robert, seem to be quite keen on school districts being keen on ChromeBooks.

    OK then, show me a ChromeBook that can’t deal with WiFi.

    This is not an abstruse argument. Cat-V, Cat-6, I don’t actually care. Normal people in normal environments all over every single use case you can think of in the civilised world in 2016 expect to be able to move their “thin client” (for want of a better word) around the “campus” (for want of a better word).

    You can do this with ChromeBooks and a proper server.

    You can’t with a Cello, as far as I can see.

  82. oiaohm says:

    3. Every cable is properly laid in a cable wall channel, or students will constantly step on and stumble on cables, causing all kind of damage to equipment.
    Deaf Spy I have been to a lot of schools I can tell you duct tape is popular. Yes roll of duct tape and stick the cable to the wall and ceiling. Completely solving that problem of students step/stumble on cable sand solved another problem. Teachers love reordering their rooms so this means you cable in wall channels and port over and over again end up in the wrong place and you have to duct tape the cable across anyhow. So might as well go with duct tape in the first place and save a few dollars and be able to reorder to the teachers preference layout. Yes teachers push for wifi so they can have preferred layout then complain like hell that they don’t have the network performance they had when they had wired. Then you end up at duct tape as they say duct tape can do almost anything bar look good doing it. Yes it looks wrong having a network cable connection box duct taped to wall at first but you get over it. Proper cabling does not mandate wall channel for network cables only that the cable is secured by something and duct tape passes all countries standard for indoor usage. At least these days you can buy duct tape is colours close to wall colours so it does not look as bad as it use to.

    Volume of cable kinda explains why a lot of deployments are ending up with a switch per room with 2 cat6A cables running 10G for a max of 55 metres of course the 10G links will be in proper channels.

    4. With all this in mind, the total length of cables will easily exceed one mile.
    True could exceed one mile if you don’t use duct tape at all. That is not exactly a problem every single time. Something like a pinet group in a class room you are talking less than 100 metre role for everything in that group mainly because that is a duct tape to desk setup. Length of cable is quite a factor.

    5. The labor to do this costs way more than any good Wifi.
    Up front labour costs can be more but the on going costs of wifi performance problems in fact end up costing more. Having people complaining about network perform issues and having to investigate over and over again that wifi causes is not worth it. Wifi you make a upfront saving and you pay for it long term more than using cable. TCO compare between wired and wifi the wired wins.

    Upfront costs do remember wifi routers are quite a bit more expensive per usable connection than wired switches. So it comes down to lengths of cable used if wifi is cheaper in up front deployment or in fact more expensive in upfront deployment. Stand alone segments like the pinet items due to lengths of cable used wifi would in fact be more expensive. Remember teachers like reordering things. So some segments that sit self contained duct tape to desks can be a good thing for happy teachers and only minor annoyed IT support after the 6 move of the segment in a year. At that point why you start understanding why using cable wall channels in class rooms is a path to headaches with unhappy IT support and unhappy teachers. Wifi performance is why teachers end up unhappy with Wifi.

    Sometimes the correct answer is horible hack in the middle in this case colour duct tape matching walls, ceilings and desks and duct tape the internal room cabling in.

  83. oiaohm says:

    The wifi ceiling is a reason why chromebooks, Linux laptops, ipads and android devices are more popular in schools than stuff running windows in portables.

    Yes another form of grid limitation.

  84. oiaohm says:

    At this moment, the imaginary teacher, unless totally mathematically challenged, would just buy 20 fanless Atom PCs with built-in 2.4 / 5GHz WiFi adapters, and a decent a/c/n wireless router, and be done with it.
    Deaf Spy. There are a few reasons why that does not work.
    http://pinet.org.uk/articles/faq.html
    Reason why pinet does not use wifi is true lack of bandwidth. There is not enough bandwidth in wifi channel to have 20 machine performing well without consuming multi channels.
    https://www.bufferbloat.net/projects/make-wifi-fast/wiki/
    Next is when you do deploy a wifi router with the modifications so it can take use most of the poor bandwidth with multi machines windows takes a hissy fit due to not handling out of order packets. Linux and OS X both can work with optimised wifi routers.

    Dear, dear… Robert, do you imply that a set of 20-30 workstations should be connected by combining five or six 8-port routers?
    That what Robert suggested performs better than wifi as long as the switches are gigabit even stacking 8 port switches. Even if you deploy wifi five or six channels routers for 20-30 workstations the shared bandwidth performance is bad and wrong. 20-30 is 5-8 channels windows.

    A good performing wifi router for Windows can only take a max of 4 computers per non overlapping wifi channel. So five channels=20 Windows computers. Due to better network stack in Linux(and Linux with right wifi card) and OS X five channels equals 40 workstations with optimised routers same speed. Its a two to 1 difference. Even so the network performance is still way under wired in.

    Please note I said wifi channels because wifi routers may have more than one channel used per device. You will run out of channels consuming them at that speed at about 200 computers using all of 2.4Ghz and 5Ghz. 25 wifi channels is a hell of a lot to use and 50 is impossible.

    If you don’t care about overlapping 5Ghz has only 25 channels in old FCC regs and only 37 FCC regs and 2.4Ghz only has 12. Remember a percentage of channels will not be usable due to noise. So using well performing wifi you fairly much have a 200 computer/device ceiling if it Linux or OS X and a 100 if it Windows.

    And those same atom powered computers would even have better support for the FOSS apps that counted than the Odroid – a win-win situation for any teacher who wasn’t a blind anti-microsoft/commercial software bigot.
    Close Wizard Emeritus. You have missed what deafspy suggested is broken because it includes wifi that is no removeable and lacks physical off switch. So similar atom powered computers without wifi and wired in just like the Odroid would be can be more flexible due to liking JIT programs more so browsers like chromium and firefox run happier.

    One of the problems with wifi is depending on the device you can have every time machines are booting up them probing the wifi network attempting to find PXE if that its turned on. Even if the machine once booted never connects to wifi so causing wifi performance issues when ever someone reboots the machines. So it a desktop unit you want removable wifi or physical kill switch that the device Deafspy suggested does not have so you can be sure it turned off if you are deploying that unit on cable. Same atom cpu with built in wifi from same vendor with wifi off jumper is 50 dollars more. Yes they know they can get away charging 50 dollars more of the off jumper due to the network troubles not turning it off can cause. Yes a 1 cent removable jumper added in instead of fixed jumper and they charge 50 dollars more.

    http://mrpogson.com/2016/10/06/amazement/#comment-353983
    Deaf Spy do go down and read this comment and work out your Supercomputer question was in fact wrong.

  85. Deaf Spy says:

    As for Oregon… One college has one room with 1 server and 24 seats… Real achievement! 🙂

    Hey, what about supercomputers? Did you manage to get the difference between a server and a supercomputer? 🙂

  86. Deaf Spy says:

    One can buy 8-port gigabit/s switches for ~$30 and cables are a few dollars each or can be made for ~$1.

    Dear, dear… Robert, do you imply that a set of 20-30 workstations should be connected by combining five or six 8-port routers?

    And how do you see how a “few dollar each” cables will do the job? A proper cable infrastructure will imply:
    1. The switch is in a box in one location of the room.
    2. As minimum, one cable leads from the box to every workstation.
    3. Every cable is properly laid in a cable wall channel, or students will constantly step on and stumble on cables, causing all kind of damage to equipment.
    4. With all this in mind, the total length of cables will easily exceed one mile.
    5. The labor to do this costs way more than any good Wifi.

    That’s not true and not what I wrote.

    Having memory lapses?

    Robert, your idea of IT infrastructure is obsolete, inefficient, and not working. No wonder no one in the world is following your example.

  87. dougman wrote, “show me one example that is currently being deployed in the U.S”.

    Well, you can zoom in on the LTSP map in USA to find things like “using LTSP since 2013” and such…

    Here’s a comment made of Southwestern Oregon Community College: “Our LTSP server is a virtual machine running Ubuntu 14.04 headless. The server is an ATOM cpu based low power server with an ISCSI target for a harddrive. Virtual machine is on another ISCSI target and user files are located on an NFS share. All using LDAP for user signon. Performance is amazing with Pentium D clients and still amazing the students how responsive a diskless machine can really be.” They’ve been running LTSP since 2014…

    So shut up about stuff you know not, dougman.

  88. dougman wrote, “LTSP appeasrs to be dead”.

    Not true. It’s alive and useful and used. See Debian.

  89. dougman wrote, “Show us one meeting that occurred in 2014, 2015 or even 2016. Admit it, there are none!”

    The software is pretty well perfected. See, it’s in Debian’s repositories. The bug reports are zero in number. It works for people. Bugs, if any, may be found in the associated packages: isc-dhcp-server, tftpd-hpa (one bug from 2014), ltsp-client (few bugs), ltsp-server (few bugs)…

    It’s a little complicated but folks can boot Odroid-* via NFS too. It’s very easy to boot from SD-card and put the NFS stuff in /etc/fstab with X -query serverip etc in /etc/rc.local on the new root file-system.

  90. Wizard Emeritus says:

    “and he is certainly in a position to know!”

    Your tinfoil hat needs adjusting…

  91. Wizard Emeritus says:

    “M$ spurred adoption of LTSP in USA back in the day…”

    And here we are 14 years later and in a different world completely, but Robert Pogson keeps harping on about the increasingly distant past as if it was the present. and toutes his amateurish hacked up mess that he came up with for solution as if it were gold.

    Nothing new to see here.

  92. Wizard Emeritus says:

    “At this moment, the imaginary teacher, unless totally mathematically challenged, would just buy 20 fanless Atom PCs with built-in 2.4 / 5GHz WiFi adapters, and a decent a/c/n wireless router, and be done with it.”

    And those same atom powered computers would even have better support for the FOSS apps that counted than the Odroid – a win-win situation for any teacher who wasn’t a blind anti-microsoft/commercial software bigot.

  93. dougman wrote, “Over 57% of deployments are located in Greece. Isn’t it a wonderful coincidence that Greece is broke.”

    • LTSP is just about the least expensive way to provide IT to a school. I know. That’s why I promote it.
    • LTSP has been rolled out widely in schools long before 2013. My biggest deployment of GNU/Linux used it and that was back in ~2006.

    M$ spurred adoption of LTSP in USA back in the day by sending out a letter demanding an audit of software. The schools were paying ~$42 per machine per annum to M$ and the audit broke the camel’s back. I set up my first LTSP lab when LTSP was sponsored by RedHat a few years later. All I did was burn a CD-image and boot the machines PXE from the server. At first I used my personal PC as the server and it ran 30 simultaneous sessions in 1.5GB RAM and later the students and I built a proper server. Total capital outlay was ~$1500 CAD for 30 clients. We already had a free gigabit port on a switch. Bam! Lose ’98 was replaced on a weekend by an inexperienced (with LTSP) teacher. Later, I figured out how it could be done in an hour with a bit of parallel processing. By gathering up the MACs while the server was being installed of GNU/Linux, that’s all the time it takes an experienced teacher to set up LTSP. For newbies there’s a learning curve but compared to lack of a budget/seats that’s a trivial problem.

  94. Deaf Spy wrote, “defunct project LTSP”.

    Nonsense. LTSP is part of many distros now and is no longer needed. The Linux kernel will boot disklessly, DHCP works, as does NFS. The only thing specially needed to boot disklessly is TFTP and it is out there. It does take a bit of effort to set up LTSP-like thin clients without an LTSP package but it can be done. I’ve done it. Unless you can safely distribute the root file-system of the server to all clients, one needs a chroot of a client-only root file-system and pass it to the clients by DHCP and NFS. TFTP passes over the desired kernel which could be custom-made to fit the clients with no extra code/drivers to slow down transfers. One can also boot locally as usual and connect to the server just with X. LTSP was set up to enable teachers with little or no knowledge of Linux or networking to do the job. It was very helpful but now distros do the job so there’s no need for a special project.

    LTSP: “You do not need to download anything from this website if you want to use LTSP.
     
    LTSP is directly integrated into various Linux distributions. You can use the package manager of your favorite Linux flavor to download LTSP. The Linux terminal server software is available in the distributions below.
     
    Debian Gentoo OpenSUSE Ubuntu “

  95. Deaf Spy wrote, “Robert doesn’t need WiFi, hence the teacher doesn’t need it.”

    That’s not true and not what I wrote. Some schools, particularly older ones, may not have any cabling except TV/video and may need WiFi. Newer schools should have built-in cabling to Cat-6. One can buy 8-port gigabit/s switches for ~$30 and cables are a few dollars each or can be made for ~$1. That’s probably less than a WiFi access port per classroom. If the construction allows one WiFi access port to supply a few classrooms, that may or may not be a better option as far as capital cost. I’ve never seen WiFi that gave better performance than Cat-6 cable. For thin clients, cable is better. For thick clients WiFi may be good enough. Many schools are ~100m in length so WiFi will need multiple access points and cable may need a central location. Only the smallest schools could get by with a single WiFi access point. The school I was in that used the most WiFi had 3 access points until someone stole one… It still had one cable per classroom and most outlets worked.

  96. ram says:

    With respect to your comments on Trump, remember this:

    “Mushroom clouds making their appearance depends on the outcome of the US presidential election. Americans are an insouciant people and do not understand the stakes. Hillary has promised conflict with Russia. Trump says he sees no point in conflict with Russia. This difference is the only important issue in the election.”

    From Paul Craig Roberts

    PaulCraigRoberts.org

    and he is certainly in a position to know!

  97. Deaf Spy says:

    Let’s elaborate a bit more on this:

    Teachers use… versus We don’t need…we have no need… TLW uses…

    and add “We don’t need/want WiFi… to the mix.

    In an imaginary classroom there is an imaginary teacher who decided to follow Robert’s wisdom and buy 20 Odroid C2s for workstations. Now, Robert doesn’t need WiFi, hence the teacher doesn’t need it. To get some connectivity, the teacher need to purchase a rather good switch and 1 mile of cables to properly setup the working environment. Result: extra costs for the cable and serious number of hours of manual labor to get the cables laid properly and tested…

    At this moment, the imaginary teacher, unless totally mathematically challenged, would just buy 20 fanless Atom PCs with built-in 2.4 / 5GHz WiFi adapters, and a decent a/c/n wireless router, and be done with it.

  98. Deaf Spy says:

    Without your (now, happily retired) beneficent hand of guidance, quite how many teaching establishments follow your nostrum of Thin Clients For All?
    Oh, wait, I know. Ecuador. Or maybe downtown Cleveland.

    Right on spot, dear Doctor. Robert’s example are Greece and a seemingly defunct project LTSP.

  99. oiaohm says:

    Show us one meeting that occurred in 2014, 2015 or even 2016. Admit it, there are none
    This is that dougman cannot read well enough.
    The meeting was not in person but was in freenode #ltsp-meeting in 2013
    The main freenode #ltsp is fully log.
    http://irclogs.ltsp.org/
    Simply because unless there is some major dispute the developers of ltsp just informally hold meetings in #ltsp. So if you go back though the 2014,2015 and 2016 irclogs of #ltsp you find informal meeting after informal meeting with them not getting into a dispute. Ltsp has formal meetings when they are not getting along. Last 3 years they have been getting along quite well. This is why looking at the changes submitted to Ltsp also in some of the submit comments mention the informal meetings. So you have found a scrap and gone off in a complete invalid direction.

    http://pinet.org.uk/
    Looks rather useless. Three blog entries, no wiki or forum. They ask people to subscribe t a mailing list! LOL… what is this, the 90s??
    Idiot strikes again the site is design for primary school student access.
    https://github.com/PiNet/
    If you want to report a issue its go to where the source is stored and use the bug reporting system. If you want to change the documentation you submit a git patch for the documentation section of the webpage so child safe way of doing a wiki so that all edits showing to primary school students have been vetted.

    This is all status normal for software targeting primary school students not to have wiki or forum or open mailing list. Just you are too big of an idiot to know this. Yes the blog not having many entries that is also common.

    Over 57% of deployments are located in Greece. Isn’t it a wonderful coincidence that Greece is broke.
    Due to Greece having departments under financial supervision they are required to report they are using stuff. Its not just Ltsp that since the Greece got the IMF bail out in 2010 we have found out about a lot of open source that was used that was never reported due to the IMF rules. So in 2010 Linux items that has been used since 1997 in Greece finally reported to the outside world. I would say until 2010 we only knew about 0.5 percent of the open source Greece was in fact using. Before 2010 Greece was 0 on the ltsp chart even that we know before the bail out there was over 100 deployments of just ltsp since they have had to report. Question what would the USA look like if reporting was mandated. Yes there has been some growth in Greece that seams to be due to debt but they were large open source users before the debt hit them.

  100. dougman says:

    “http://pinet.org.uk/ Welcome to hybrid thin-thick client for PI boards. Currently used in the USA.”

    Looks rather useless. Three blog entries, no wiki or forum. They ask people to subscribe t a mailing list! LOL… what is this, the 90s??

    HAHAHAHaaaahha.

  101. dougman says:

    “What about it, show me one example that is currently being deployed in the U.S.
    Smart ass question really. ”

    Well, as I was taught long-ago. Rather be known as a smart-ass, versus being know as a dumb-ass. You dear one, fit the peg hole on the latter one perfectly.

    By the way, that pie chart is interesting.

    http://www.ltsp.org/stories/widget-countrystats/

    Over 57% of deployments are located in Greece. Isn’t it a wonderful coincidence that Greece is broke.

    https://www.theguardian.com/business/2016/may/09/greece-financial-emergency-tragic-absurd-imf-debt-relief

  102. dougman says:

    “Seems the last meetings occurred back in 2013. So what progress has been accomplished in the past three years? NONE! – So this is not true the changes list show that is not fact but your normal made up garbage based on your incompetence at doing research. ”

    Show us one meeting that occurred in 2014, 2015 or even 2016. Admit it, there are none!

    http://wiki.ltsp.org/wiki/Meeting:Upcoming

  103. oiaohm says:

    Go to you page you quoted dougman and see the word changes click it. Then admit the project is still actively changing things.
    http://bazaar.launchpad.net/~ltsp-upstream/ltsp/ltsp-trunk/changes

    dougman if you had followed direction you would have been at that link. Apparently you cannot read.

  104. dougman says:

    “dougman go find you glasses or learn how to read. release.conf 2752 2016-09-06 16:45:06 UTC”

    Ok sure, lets see.

    http://bazaar.launchpad.net/~ltsp-upstream/ltsp/ltsp-trunk/view/head:/release.conf

    Whats this? OH!..its a configuration file. Which is just a simple text file. What’s in it??

    1 NAME=ltsp
    2 VERSION=5.5.8
    3 EXPORTCHANGELOG=yes

    Three lines you dweeb!

    LMAO.. see, you think a simple TEXT file is updated it is current software. HAHAHHaa.

  105. oiaohm says:

    http://pinet.org.uk/
    Welcome to hybrid thin-thick client for PI boards. Currently used in the USA.

  106. oiaohm says:

    dougman go find you glasses or learn how to read.
    release.conf 2752 2016-09-06 16:45:06 UTC
    LTSP was only version updated in 2016 so is not dead yet.
    Go to you page you quoted dougman and see the word changes click it. Then admit the project is still actively changing things.

    Seems the last meetings occurred back in 2013. So what progress has been accomplished in the past three years? NONE!
    So this is not true the changes list show that is not fact but your normal made up garbage based on your incompetence at doing research. Yes LTSP is in maintenance mode so there has been some progress. Will be at some point superseded by multi-seat support work allowing full opengl rendered application output to be sent to clients.

    https://www.nomachine.com/terminal-server
    The reality is the closed source terminal server on Linux does support opengl applications on the thin-clients.

    As the cpu in thin clients are getting more powerful using http://www.drbl.org/ is also coming more popular.

    What about it, show me one example that is currently being deployed in the U.S.
    Smart ass question really. Currently deployed you can give numbers on. Currently being deployed it read the LTSP mailing list and hope to spot clues. There are quite a few currently being deployed in the USA new deployments from 2015 got reported in 2016.
    http://www.ltsp.org/stories/ There are 4 on that page if you spend time to find them dougman in the USA in schools new in 2015 and reported in 2016. I will say ltsp has the worst interface ever for going though success stories. Also nothing odd for success stories not to be updated for 4 years then have a update saying yes we were in fact using ltsp for the past 4+ years.

    When a product does not have required subscriptions counting how much is being deployed is hard.

  107. dougman says:

    To continue further, LTSP appeasrs to be dead.

    http://bazaar.launchpad.net/~ltsp-upstream/ltsp/ltsp-trunk/files

    Seems the last meetings occurred back in 2013. So what progress has been accomplished in the past three years? NONE!

  108. dougman says:

    “See LTSP”

    What about it, show me one example that is currently being deployed in the U.S.

    “Oh, and Chromebooks are essentially thin clients. Last time I checked Google did not use a lot of TOOS servers.”

    WOW Robert, that’s a large stretch of the imagination. I dare you to say take a thin client to McDonald’s or Starbucks, and try to post an entry on your blog here. Whats that you say? No working??

    “Hundreds of schools in Greece have recently made the change to LTSP, sometimes using TOOS as their thin client platform, but the server runs Ubuntu GNU/Linux.”

    Well, Hammy did mention “Schools in USA” and that was the juxtaposition of my point. But since you brought up this Greece school, I see they only setup six clients. Doesn’t sound like a winning implementation does? I mean this one off setup wouldn’t happen to be the outcome of Greece going bankrupt would it? Just a coincidence??

  109. oiaohm says:

    https://www.researchgate.net/publication/308786278_HeNN_LTSP_E-library_Deployment_Report_in_rural_government_school_of_Nepal_Syangha_Malunga-1

    No school is deploying thin clients these days and on top of that, when they did deployments in the past, it was a Windows server.

    If you look around you find new thin client deployments all over the place pure Linux based. Sorry the idea that thin clients run linked to windows servers was all that existed is you pushing false facts as normal.

    Chromebooks end up with schools running web servers to run moddle and the like.

    Yes it true the windows terminal server solutions are disappearing but the Linux terminal server deployments are increasing.

    These day’s it is namely Chromebooks making headway. Cheaper to procure and manage long-term for administrators.
    There are three things making head way.

    Chromebooks with Linux Webservers.
    Linux thinclients with Linux terminal services.
    Linux laptops.

    Sorry to say Linux thinclients with Linux terminal services work out cheaper to procure than chromebooks.

  110. dougman wrote, “No school is deploying thin clients these days and on top of that, when they did deployments in the past, it was a Windows server.”

    See LTSP

    Oh, and Chromebooks are essentially thin clients. Last time I checked Google did not use a lot of TOOS servers.

    Hundreds of schools in Greece have recently made the change to LTSP, sometimes using TOOS as their thin client platform, but the server runs Ubuntu GNU/Linux.

  111. dougman says:

    “Schools in USA and other countries thin clients is part of their deployment plans. ”

    No school is deploying thin clients these days and on top of that, when they did deployments in the past, it was a Windows server.

    These day’s it is namely Chromebooks making headway. Cheaper to procure and manage long-term for administrators.

  112. oiaohm says:

    Without your (now, happily retired) beneficent hand of guidance, quite how many teaching establishments follow your nostrum of Thin Clients For All?
    It would pay DrLoser yourself if you got to date on the topic.
    http://fossforce.com/2016/10/charlie-reisinger-foss-education-proponent-practitioner/
    Schools in USA and other countries thin clients is part of their deployment plans. In fact they deploying two plans normally.

    Thin clients for all and laptops for all both at the same time. Please note selected laptop to be a laptop that draws less than 100W closer to 50W the better. Locked down thin-clients are good for performing exams on students that are closed book. Laptops are good because students can take them home and work on stuff.

    Thin clients advantage is how low of power they can be yes a LCD + a 10W thin client fits in the 37.5 watt limit that you will find at the worst power points like old electronic typing teaching rooms. Electronic typewriters are not that power hungry. So we need computers that fit in the power envelope of a Electronic typewriter.

    The one thing you can say if the computer uses over 100W schools don’t want to deploy lots of them because it gets too expensive. Lower the power usage the better for the schools running budgets. There are already many schools where all that is left is low power using laptops and thin-clients.

  113. DrLoser says:

    I was a teacher. I helped other teachers obtain and use PCs. I know what I’m writing about.

    Apparently, Robert, you lack the ability to understand two rather important things:
    a) Logical deduction
    b) Bathos
    The first would get you laughed out of any theoretical position as an unqualified teacher.
    The second is … well, look up “bathos.” It just makes you laughable.

  114. DrLoser says:

    Teachers use… versus We don’t need…we have no need… TLW uses…

    Sounds like a fair challenge to me, Robert. I’m certain that you are up-to-date with all things pedagogical, what with being completely unqualified to be a teacher, and all.
    Do tell. This is 2016. The teaching world is bigger than Easterville, and (thank god) does not have an ignorant bigoted tyrant in charge.
    Without your (now, happily retired) beneficent hand of guidance, quite how many teaching establishments follow your nostrum of Thin Clients For All?
    Oh, wait, I know. Ecuador. Or maybe downtown Cleveland.
    Go on, impress us with your up-to-date knowledge on the subject.

  115. Deaf Spy wrote, “Extrapolating your small personal needs to teachers and then to the whole world is just silly and laughable.”

    I was a teacher. I helped other teachers obtain and use PCs. I know what I’m writing about.

  116. oiaohm says:

    Perhaps I should ask our dear old Robert whether a distributed computing unit of thousands of personal computers and smart devices can be considered a supercomputer.
    http://phys.org/news/2015-08-supercomputer-palm.html
    Deaf Spy 2015 it was theoretical.
    http://www.cnbc.com/2016/05/18/fight-zika-by-making-your-smartphone-into-a-supercomputer.html
    2016 its reality. So its not a question any more.

  117. Deaf Spy says:

    Dear Doctor, it became obvious that not only our doesn’t dear old Robert perceive the server roles and the part CPUs play in every role; he doesn’t even realize what a supercomputer is.

    Perhaps I should ask our dear old Robert whether a distributed computing unit of thousands of personal computers and smart devices can be considered a supercomputer. Alas, I don’t believe such a question will prompt him into an alternative lane of thinking.

  118. Deaf Spy says:

    Teachers use… versus We don’t need…we have no need… TLW uses…

    Robert, why don’t you get your ideas straight and stop moving the goalpost from your little personal needs to teachers, to the world and back? Pick one and focus on it. Extrapolating your small personal needs to teachers and then to the whole world is just silly and laughable.

  119. oiaohm says:

    Deaf Spy current photoshop and autocad does not run on a atom cpu as extensions those programs used are not in atom cpus. Items like SSE is missing. But as a thin client box its decent. So quite a few win32 and win64 applications will turn up toes and die when attempting to run on Atom x5-Z8300.

    Please note the machine you quoted Deaf Spy fits in a 25Watt power limit. dougman said Repeat after me, “IS TOTALLY USELESS” as 25W cannot power much of anything.
    You can do quite a lot inside the 25 watt limit.

    To be truthful for a thinclient that is going to be wired into the network
    $15 for a cheap WiFi dongle
    You are not going to be doing that in the thinclient case. Built in wifi and bluetooth can increase radio noise. You don’t want 20+ computers in room having active wifi or bluetooth that are not connecting to the wifi network.

    Odroid C2 it self is only max 10Watts but most of the time it never gets above 2.5Watts.

    Having USB 3 pushing your required power up. USB 3.0 mandates 900mA power min vs USB 1 and 2 that are 500mA min per port.
    http://openbenchmarking.org/result/1608052-HA-1608034HA54
    You find the same Atom x5-Z8300 on the Up Board 2G board that is a PI clone and it has 4 USB 2.0 instead of 2 USB 2.0 and 1 USB 3.0.
    http://openbenchmarking.org/result/1603300-GA-1603301GA94
    Odroid C2 is not that bad performing when you are doing watt vs performance yes its slower than the Atom x5-Z8300 but the Atom x5-Z8300 is eating more power. Performance per watt the Atom x5-Z8300 and the C2 cpu are about equal.

    But then there is the elephant in the room C2 cpu is 28nm and the Atom x5-Z8300 is 14nm it should be running rings around the C2 in power effectiveness but it does not.

    Odroid XU line is the Odroid line built to 20watts instead of 10watt that the Odroid C lines are Odroid XU currently has not been updated to Arm64 yet. When Odroid XU line is updated will be interesting to revisit. Deaf Spy basically building to a watt limit comes at a price and a lot of PI and PI Clones are built to quite low watt limits. At times that watt limit is critical particularly if you are going to stick it on a robot and run it from battery.

    Modular nature of the PI class enabled their power profile to be changed to suite what they are currently being used for.

  120. Deaf Spy wrote, “Robert can’t tell a super computer from a server.”

    Well, a super-computer is essentially a cluster of servers. They can run Desktop GNU/Linux if they want… X uses the network and super-computers usually have that covered. One thing a super-computer likely isn’t. That’s a client PC.

  121. Deaf Spy wrote, “Keep dreaming, Robert, keep dreaming.”

    We don’t need/want WiFi so pare off $15 and our SD card cost ~$7 CDN so cut a few more dollars. Also, we have no need for PS or AC. TLW uses NFS from Beast III. So, cut the crap.

  122. Deaf Spy says:

    Sad to see how our dear old Robert can’t tell a super computer from a server.

  123. Deaf Spy says:

    Teachers use what works and ARM is far more affordable than x86.

    Reality check.

    Ordroid C2 costs $40. Comes with no storage, no WiFi module, no Bluetooth, no antenna, no USB 3.0 support. To use it, you need to spend at least $10 for a cheap SSD card (slower), $15 for a cheap WiFi dongle, and you can forget of USB 3.0. Of course, you need to make sure the dongle works with Odroid and is supported by whatever distro you put on it. Total: at least $65

    Z83 costs $95. You get everything the bare-bone Odroid gives plus:
    – 32 GB eMMC storage.
    – Built-in WiFi and BT support, and an external antenna.
    – USB 3.0 support.
    And it runs all the software in the world, including Photoshop and AutoCAD.

    Keep dreaming, Robert, keep dreaming.

  124. oiaohm says:

    2400W is basically the figure you start with for a power point.

    100W per point that is 23 seats allowing 100 watts for switch operation and loses. 55W is about 40 seats
    25Wx2 POE is roughly 50 watts using two current PoE cables to regulator board is still about 40 seats per 2400W supply with the possibility of doubling the seats in future by upgrading switch and clients so having to change no cabling.
    220W for a PC is 10.

    Those are all the general per standard power point circuit number of seats.

    Lets say we do a seating requirement of 200 students seats to use computers all at the same time.
    20 circuits to have 200 220W PC.
    10 circuits to have 200 100W level solutions.
    5 circuits to have 55/50W level solution.
    Of course you could say the server room needs 2 circuits to power the servers. At the 55/50W level you are looking less than half the power consume of a normal PC even running the servers. Chromebooks land in the 100W level without a server is still half that of a quite poor PC.

    Doubling the circuits you have to install does increase expenditure a lot on deployment as does increase on going.

    As you said capital funding is a issue paying the power bill eats into future capital funding. Power grid and power usage are factors. How much power grid the solution will need and what the power bill will look like after deploying the solution are things schools and other places running massive number of machines are worried about.

  125. oiaohm says:

    Repeat after me, “IS TOTALLY USELESS” as 25W cannot power much of anything.
    https://www.pi-supply.com/product/pi-poe-switch-hat-power-over-ethernet-for-raspberry-pi/
    25 watt can fully power all the existing pi and most of the clones with watts left over. So not useless in thin client set-ups. 50W does give room to power screen.

    So if you have 37.5W limited power point and want a computer at that point the 25W on PoE can be quite useful to run thin client so the 37.5W can be given fully to the LCD screen.

    This is where thin client systems start coming into their own. Of course new standard 55W and 100W will be more practical. Really able to run a thin-client seat on one cable.

    Dougman you are the one who is being totally useless. You would ignore usable resources. Something else interesting is PoE does not interfere with network packet transfers but the 50/60hz from normal AC does. So getting thin-clients away from 240/120 AC does in fact improve their performance.

    Have you seen the one-line diagram? If not, then you are speculating and just wasting your breathe.
    I know the regulation limits. Yes I have seen line diagram. I am not speculating.

    LOL… actually, capital funding is the major hurdle. A school is not a data-center you dweeb.
    Both are capital funding and power grid issues are linked. Having to alter the power grid in the school equals increased cost and only can be performed at particular times. Grants that provide capital funds have time limits. So having to alter power grid can result in less capital fund access due to grant time limit being crossed before operationally deployed. Yes grants are evil that if you don’t meet the time line all money technically has to be returned. Grant access having to submit time lines and stick to them.

    Basically you incorrectly guessed as normally. Amount capital funding and power grid inside schools is linked worse the schools power grid the lower the funding access they deploy computers. Kinda completely backwards to the way it should be.

  126. dougman says:

    “Power over Ethernet”

    Repeat after me, “IS TOTALLY USELESS” as 25W cannot power much of anything.

  127. dougman says:

    “The biggest barrier to deploying computers in schools is the power grid of the school.”

    LOL… actually, capital funding is the major hurdle. A school is not a data-center you dweeb.

  128. dougman says:

    “What you got incorrect is a room having two points does not mean those two points are on two circuits. So you cannot double because you have two points because most likely they are 2 points one circuit.”

    Have you seen the one-line diagram? If not, then you are speculating and just wasting your breathe.

  129. oiaohm says:

    https://en.wikipedia.org/wiki/Power_over_Ethernet
    I did not mention really the worst. Power over Ethernet currently does 25.5 W of power so two cables per seat and you can run something like PI/clone based with screen. This makes a power point at 37.5W look good.

    IEEE 802.3bt upcoming will allow 55W to 100W per seat just one the network cable. This is per cable from poe switch.

    This is the problem. Power per seat. 200W PC is way too much for how much power you can be sure to get to a seat. The target need to be at least less than 100W. Lot of chrome-books come in under 65W.

    Some of the reason why schools consider some chromebooks models more than others chromebooks models is their power usage compared to their schools power grid limits. Power grid is a huge problem.

    Yes there are two common possible power grids in a schools. PoE power grid and the AC power grid. Both are limited on amount of wattage that is usable. PoE is more predictable due to its per port nature.

    Understanding what the wattage limits are explains why Chromebooks and PI/clones are doing so well in schools.

  130. oiaohm says:

    dougman
    Oh? Sans looking on Wikipedia, do instruct me on my erroneous command of Ohms law.
    You got your Ohms law right non qualified get that right as it taught in school. What you did not get right is the number of circuits. 2400Watts max out a single point is correct. What you got incorrect is a room having two points does not mean those two points are on two circuits. So you cannot double because you have two points because most likely they are 2 points one circuit. Also does not mean those 2 points are on the circuit alone. This is basically electrical distribution. You will find up to 32 double point plates on per circuit (64 power points)it does not matter if it a 120V 20amp in Canada or Australian 240V 10amp so the most horible location you can find yourself in is 37.5 watts per point of usable power.

    20 watts to run a screen is not that avoidable. So you have approx 15 watts to play with in the worst case.

    The biggest barrier to deploying computers in schools is the power grid of the school. Power grid being how the power circuits were installed in the school and how little usable wattage per power point is in fact at each one has if you wish to use all points.

  131. dougman says:

    “Also 120V 20amp power point cable as a 120V 20amp fuse in most cases”

    What is a power point cable? Why did you repeat “120V 20amp” twice in the same sentence?

    “As you said 120V 20 amps 10 PC class computer max. Building in a school might have 16 rooms 2 points per room one circuit. Little problem here in this case is 2400 watts/32 so only 75watts per existing wall point to play with use any more and other rooms will not have usable power.”

    No clue as to WTF you are attempting to convey.

    The issue here is a double power-point on wall only has 1 cable and 1 fuse behind it. Just because you have two power points does not double the amount of watts you can use compared to a single power point. Its only double if you have two circuits. Power point ratings are based on if you turn off all other power points on the circuit you could possibly draw that otherwise you have to use less.

    Ok, I am seriously interested. Not only do I NOT know what a power point cable is, what is a double power-point? Hmmmm…power points have ratings now? Is this a new function of Microsoft Office?

  132. dougman says:

    “Sorry it attack my IQ again”

    ?

    “you were pulled up for being completely wrong again.”

    Oh? Sans looking on Wikipedia, do instruct me on my erroneous command of Ohms law.

  133. oiaohm says:

    Also 120V 20amp power point cable as a 120V 20amp fuse in most cases. As you said 120V 20 amps 10 PC class computer max. Building in a school might have 16 rooms 2 points per room one circuit. Little problem here in this case is 2400 watts/32 so only 75watts per existing wall point to play with use any more and other rooms will not have usable power.

    The issue here is a double power-point on wall only has 1 cable and 1 fuse behind it. Just because you have two power points does not double the amount of watts you can use compared to a single power point. Its only double if you have two circuits. Power point ratings are based on if you turn off all other power points on the circuit you could possibly draw that otherwise you have to use less.

    Dougman anyone with electrical knowledge should have know you cannot just stack on power boards to get more points. Its how to overload power distribution circuits.

    So there is no way I believe you have proper electrical skill. If you have managed to get qualified someone passed you when they should not have.

  134. oiaohm says:

    Arguing over volts, amps and watts with someone that’s being in the electrical field all his life, just laughs at your blathering.
    Then you should have known that on old electrical circuits that more power-points were put on the feed cables than the cable could take because it was believed they would not all in be in use at the same time.

    Sorry it attack my IQ again because you were pulled up for being completely wrong again.

  135. dougman says:

    Spammy, don’t try to pass yourself off as intelligent, as your are not.

    Arguing over volts, amps and watts with someone that’s being in the electrical field all his life, just laughs at your blathering. You put the name of “ohm” to shame; perhaps I should trademark your username and send you a C&D for TM infringement. Hmmmm…decision..decisions.

    I would classify you as the worlds dumbest man. You, yes you, who poses no more of a IQ then the world’s smartest termite. I don’t blame you though, really I don’t, intelligence is namely based on good genetics and educational learning, neither of which you have apparently.

  136. oiaohm says:

    In addition, the nominal circuit power to each outlet would be 20-amps. Canada has a socket voltage of 120V at 60Hz. Using simple math, this gives you 2400 watts per outlet.
    Dougman defective maths. In old builds all the power points will be on on circuit. So the wall limit is based on how much power the cable and fuse to the power points can provide not how much the wall point itself can take.
    Using simple math, this gives you 2400 watts per outlet.
    So this simple maths is insane optimistic and if you used that in a few rooms in building you can fairly much be sure to blow a fuse or burn cable in wall. The max PC computer(box and screen) per power point you should consider is 8 into a 120V 20watt rated power point. This is allowing for a PC power supply malfunction resulting in a blown fuse not something worse. So no way 10 per point.

    School normal forbid teachers going nuts with AC power-boards some schools to as far as totally forbidding AC power-boards to prevent overloading the cabling in wall. Schools policy mandate that everything must be connected properly to prevent wall fires and blowing fuses. So dougman your 20 computers per classroom is just you being electrically clueless. This should not be a surprise to anyone any more.

    So let me get this straight. Running extension cords is a no-no, but but running wires all over the place from said 12V power supply, which would have to be fairly sizable is a better solution? Where is the logic in that??
    There is a stupid reason for this. Schools a long time ago approved running model railways as safe for students at up to 20VDC at 15A. Please note the DC bit AC interferes with muscle control so where you get zapped by DC and you let go. AC you can find yourself unable to let go so get a lot more power passing though body so way more likely to die.

    So using a 12DC rail at under 15A is allowed for teachers to deploy because it just deploying a model rail without trains. You will notice most charging bricks for Chromebooks the DC part is less than 20VDC at 15A mostly because this is the max that all students students are allowed to handle. So a 120VAC is something only teachers/responsible adults are meant to touch in quite a few class rooms.

    So deployment rules of schools are a pure pain in but.

    “extension cords” and “power strips” if they are DC they are a lot more usable in schools. The issue is you are suggesting using AC versions dougman.
    http://www.rpc.com.au/catalog/2-pin-polarised-wall-socket-p-1075.html
    That is the student safe wall socket that teachers can allow all students to plug appliances into. There are 2 Pin Polarised extension cords and power strips. So 120VAC to 12VDC to DC extension cords and power strips is deployable where 120VAC to 120VAC extension cords and power strips in a lot cases not deployable.

    In dc you have a 480W limit and that is if you allowed to deploy 32Volt if you are only allowed to deploy 12Volt you have 180W limit. So Odroid-C2 x 6 is hitting limit. So 200 watts per computer and 20 watts per LCD screen this has in fact eaten you out of house and home if stuck in a 12volt system. Odroid-C2 you can deploy 3 with LCD screens inside limit of 12 volt teachers are allowed to use.

    educational Chromebooks for their students
    Some Schools are deploy PI/clones because the reality is they don’t have the power grid in the school to support students with chromebooks or PCs. Its how to give enough seats for students inside the power-budget you have got. Its fairly cheap to add a few extra power points to a room to support a servers on a decanted line is really expensive to rewrite the complete school. Remember lots of schools have insanely tight budgets.

    So both Chromebooks and PI/clones fight for exactly the same market the PI/clones trump cards is low power usage as well as being flexible for robotics but have to be deployed with terminal servers to allow max function and Chromebooks trump card is so call easy maintenance with no servers but with restricted robotics usage as well as heavier power usage. So at this stage neither one is a absolute win for education.

  137. dougman says:

    After reading your mess of a solution, it’s no wonder educational Chromebooks for their students. No need to run anything, just had out, login and off they go!

  138. dougman wrote, “running wires all over the place from said 12V power supply, which would have to be fairly sizable is a better solution?”

    Let’s see… Odroid-C2 draws 2A max. A cluster of five or six needs 10-12A, like one headlight on my car. Do you need a bunch of power-supplies to do that? I don’t think so. One battery or one battery-eliminator takes care of it. The monitors do need serious power so a power-bar makes sense but you just need one per cluster, perhaps bolted under the table. Teachers use what works and ARM is far more affordable than x86. The threshold that needs to be overcome in education is not CPU power but seats. I’ve used multiseat X and a cluster. Both work for schools.

  139. dougman says:

    “That is not a proper connection.”

    Agreed. This school sounds to be too cheap to update it’s infrastructure.

    “it is easy to distribute 5-12V to a cluster of ARMed PCs on a table.”

    So let me get this straight. Running extension cords is a no-no, but but running wires all over the place from said 12V power supply, which would have to be fairly sizable is a better solution? Where is the logic in that??

    “Extension cords are intended for temporary use.”

    As if your solution is a more permanent one? LOL..

    Perhaps people should start using wireless electricity, as Tesla imagine and almost achieved a hundred plus years ago.

  140. dougman wrote, “there are things called “extension cords” and “power strips” that will give you multiple outlets.”

    That is not a proper connection. However, it is easy to distribute 5-12V to a cluster of ARMed PCs on a table. Extension cords are intended for temporary use. They expose students and the building to increased risks which are not present with low voltages.

  141. dougman says:

    “They don’t even need to add more electrical outlets because of the minimal power requirements compared to PCs. I was in one school that had only two electrical outlets in each classroom.”

    I don’t know Robert, seems like you taught at Inuit schools. Two power outlet per room? That seems feasible, but honestly, there are things called “extension cords” and “power strips” that will give you multiple outlets.

    In addition, the nominal circuit power to each outlet would be 20-amps. Canada has a socket voltage of 120V at 60Hz. Using simple math, this gives you 2400 watts per outlet. So total, that’s 4800 watts in any classroom setting. If you figure 200 watts per computer and 20 watts per LCD screen, that allows you well over twenty computers per classroom doesn’t it?

    So your premise and reasoning does not keyhole into your ideology. Each classroom could have easily supported many computers.

  142. dougman wrote, “80% of those bought are not even being used”.

    One could write the same about the latest Intel CPUs. They are idling 80% of the time. In education, a lot of equipment sits idle when the course or subject matter for which they are acquired is not current. ie. in K-12, they might be used for a couple of weeks as part of middle school or high school curriculum. Schools love them because they cost less than a PC yet teach the basics just as well or better for the hands-on stuff. I used to teach kids with ATX PCs hands-on to a great extent so I can appreciate that. Regular classroom teachers could not use PCs nearly as much as I with a lab full of them. With the Raspberry Pi, even classroom teachers can afford a cluster. They don’t even need to add more electrical outlets because of the minimal power requirements compared to PCs. I was in one school that had only two electrical outlets in each classroom. Adding any kind of computing devices was a challenge. The ARMed PC is cheap, small, and useful in schools as they are in my home.

  143. Deaf Spy wrote, “ARM on server is good only for cache and file servers, where the CPU is just an orchestrator to run the OS”.

    Well, a lot of servers fill those roles, but ARM has computing power too.

    August, 2016: “Fujitsu said this week that it will use ARM-based processors to build a successor to an existing Japanese supercomputer called Project K. Fujitsu is building the “Post-K” machine for the Riken Advanced Institute for Computational Science, which plans to use it for biomedical, climate, and energy research. The computer is slated to be installed and turned on in 2020.”

    That’s with ARM adding vector processing to its architecture which is already good enough for many tasks and energy-efficient. Sooner to arrive in the pipe of technology is the X-Gene-3 chip which is aimed to be competitive with Xeons next year. I suppose Intel may well advance another step by then but to say that ARM can’t serve generally is a lie. Even the lowly Cello can do a lot more than file/cache. Obviously, folks are using it for desktops and software development. They would not do that if it were as feeble as Deaf Spy claims.

  144. oiaohm says:

    Pi’s sold what 10M this year? Breaking this number down, 80% of those bought are not even being used, 20% are being used according to Pareto. Whereas, Chromebooks are more user friendly and far surpass any expectation from the Pi.
    Dougman. This is also shocking to look up the stats on the number of Chromebooks are collecting dust.

    The reality most PI and most Chrome-books are sold to the education market. The percentage of both the are ending up thrown in a box somewhere forgotten about once the class is done using them is about 80%.

    Since you have raised the topic of binary packages
    http://mrpogson.com/2016/10/06/amazement/#comment-353831 Don’t say mocking its not quoted. Anything not quoted by academics is yours dougman. So would you please explain how binary packages are broken. Lets test out if you know this topic or not.

    Your statement is akin to comparing a VW Beetle to say a four-door KIA. Given the option, people would choose to use the Chromebook instead.
    The reality here is most people are not willing buying Chrome-books or PI devices instead either or both is forced on them.

    Deaf Spy
    He can’t just grasp the fact that ARM on server is good only for cache and file servers, where the CPU is just an orchestrator to run the OS, and it is the RAM and IO throughput that service the actual operations.
    This is kind of true and kind of false. Saying good only is kinda over stating a bit. Cache and File server don’t normally go anywhere near JIT. If you want to know why Firefox has so much of a issue on different ARM chips look no further than the JIT. ARM64 design is pure W^X with hacks to allow WX to exist for JIT that on some chips work and on some chips don’t even mono is learning this on Android.

    ARM64 is good with two particular rules.
    1) Everything Ahead of Time complied. This is why you see ART in Android 4.4.
    2) Use co-processing and multi threading systems.

    He makes the dire mistake to put ARM on the same train as POWER, Xeon, and Sparc.
    You included 1 platform too many.
    http://www.pcworld.com/article/3086249/fujitsu-drops-sparc-turns-to-arm-for-post-k-supercomputer.html
    Deaf Spy something horible is 10nm arm64 turned out when fujitsu tested them kicked the living heck out of 10nm Sparc designed chips so the next Fujitsu super computer is ARM. So Sparc is no longer on the same train. Power and Xeon are the two ARM has to beat with ARM64 they beat Sparc at the same nanometre. Sparc only has advantage over a ARM64 chip at the same nanometre is if Sparc is like the Oracle S7 with coprocessors in the chip like the DAX offloading tasks like SQL processing. Sparc only performs ARM64 when it has a coprocessor to assist it with the task if it does not have the coprocessor for that task welcome suffering. Both arm and sparc have fabric frameworks for off loading tasks between cpus.

    So ARM64 is basically taking Sparc slot of being in third place. Sparc really should not be put in the same train as Power or Xeon either.

    Oracle to make the Oracle S7 look good against Xeon does two particular things. Use a SQL benchmark that DAX gets to help with a lot. Using JIT based tests that Xeon and all current day x86 chips in fact hate only minor less than ARM64 bit chips. Of course ARM64 hate could be reduced by formalising some valid way to perform JIT require task of updating JIT code.

    Power does not matter what you try. Power likes JIT like AOT code fairly much equally. Order of competitive problem.
    1) Power likes JIT and AOT equally.
    2)Xeon/x86 like AOT a lot hates JIT but is not going segfault because of it.
    3)Arm64 like AOT absolutely hates JIT. So bad that JIT only works with hacks particular to the Arm64 chips being used.
    4)Sparc like JIT shock hates AOT. Why hates AOT each new version of the chip normally adds instructions that programs running on that chip have to use or fail to run on that chip. So a mix of generations in Power, Xeon or Arm64(even before arm chips before arm64) Using program build for the oldest generation of the and it will run on the complete pool. Example current openpower with old powerchips the binary build for older power chips work on the new openpower chips no problems. Swap openpower /old powerchips for Xeon and arm64 equal and it works.

    https://wiki.debian.org/PortsSparc The AOT fault basically kept on breaking the sparc ports of debian until finally they gave up supporting Sparc at all.

  145. Deaf Spy says:

    With img tags obviously not supported, here is the link:

    https://media.giphy.com/media/3o7TKVDYQ9cl7lzVLy/giphy.gif

  146. Deaf Spy says:

    He only desires ARM as a thin client (which it can just about manage) and as a server (which it cannot, as yet.)

    But Robert keeps insisting that ARM is there for the desktop, dear Doctor. I wonder why? Perhaps it is because he confuses a thin client with a desktop? Or is it because he extrapolates his rather limited needs to the nothing-suspecting humanity?

    It is also obvious our dear old Robert doesn’t understand the purpose of ARM on the server market. He makes the dire mistake to put ARM on the same train as POWER, Xeon, and Sparc. Poor Robert! He can’t just grasp the fact that ARM on server is good only for cache and file servers, where the CPU is just an orchestrator to run the OS, and it is the RAM and IO throughput that service the actual operations.

    All in all, Robert’s skills illustrated:

  147. dougman says:

    So resorting to miss quoting because you cannot accept being wrong. No its possible to mock a person without miss quoting. Of course cannot like being pulled up that miss quoting is invalid so now is going to quote everything to try to get me to give that miss quoting is allow.

    This just shows the level idiot you are who does not know the difference between mocking and miss quoting. Its the old idiot excuse that mocking allows them to do miss quoting and other major mistakes.

    So lets be more of a fraud . I have not raise the fact you don’t have a clue about farm equipment or anything else. It was me who point out that the breakages robert was suffering from was avoidable by ploughing.

    You know, this is a super big maybe. I have mentioned it before but the process is a pain in ass. The fact you referring to miss quoting as mocking really means you are a idiot. You said that miss quoting is fine for mocking right.

    Please use reference sources that are not conflict of interest in future then you may not be basing you ideas fiction.

    Badly packages applications exist. Depending how built Yes it would be better at long last has agreed to sit down and work Ok it would be nicer

    Please use reference sources that are not conflict of interest in future then you may not be basing you ideas fiction.

    Interesting point. There is something interesting here. Since you were completely wrong again please do this to yourself.

    There are about 12 cases that has done this on different topics including religion. I typed that line because I am sick of the same repeated mistake. So much the 100 percent normal idiot game play.

    Since I suffer from dyslexia I have to understand the diagnostic methods used for it and the statistics about it.

  148. dougman says:

    “dougman there is a odd number here. Raspberry PI sell in the same volume as chrome-books.”

    The only oddity here is you.

    Raspberry Pi’s are not even in the same class as a Chromebook, and your statement is ignorant at best.

    Pi’s sold what 10M this year? Breaking this number down, 80% of those bought are not even being used, 20% are being used according to Pareto. Whereas, Chromebooks are more user friendly and far surpass any expectation from the Pi.

    Your statement is akin to comparing a VW Beetle to say a four-door KIA. Given the option, people would choose to use the Chromebook instead.

  149. oiaohm says:

    Wishing it go away? No one is saying or doing that, these devices pretty much do that themselves. Chromebooks are FAR more ubiquitous then any of the boards you have mentioned
    dougman there is a odd number here. Raspberry PI sell in the same volume as chrome-books. ODROID C2
    http://www.hardkernel.com/main/products/prdt_info.php?g_code=G141578608433&tab_idx=2
    Out of all its 40 pins on the GPIO header only 3 are different to what are on the Raspberry PI.

    So if you group the Raspberry PI with the Raspberry PI compatible boards(board that have the Raspberry PI GPIO 40 pin interface +-4pins) for volume sold vs chrome-books volume sold so you are doing multi-vendor vs multi-vendor chrome-books are not selling as much as people would first think.

    The Raspberry PI has started the x86-pc clone thing all over again. Just like the old PC clones some Raspberry PI clones are better than others.

    DrLoser missed something .
    https://www.bloomberg.com/news/articles/2016-08-16/intel-licenses-arm-technology-in-move-to-boost-foundry-business
    Intel backs both sides of the fence. So Intel makes x86 and ARM. Intel is not absolutely sure what architecture will win in the end.

    Intel PC business does not include just x86 chips they also include ARM chips they sell to the mobile market in their PC numbers. Because by Intel logic mobile market are personal computers. So if we came back in 10 years time Intel could be making some arch we have not even heard of yet. Intel key business is fabrication and they will license what ever it takes to be competitive. The death of x86 in use would not be death to intel.

    Hardware ARM fragmentation is killing the platform the same way software Linux fragmentation did. Unless someone gets the reigns, there will be no year of ARM on desktop, not next year, not the year after, but never.
    Deaf Spy we are seeing work with flatpak and snappy to source out the software fragmentation on Linux. If you look deeper down in the ARM platform you also see a lot of work reducing fragmentation.

    ARM themselves with the first generation of ARM64 included a standard way for the OS running on it to identify what it was in fact running on. The ARM chips prior to ARM64 had no way to-do that so kernels had to be made matched exactly to the old arm 32 bit chips.

    So it the issue hardware fragmentation no so much ARM has addressed most of it. Secureboot caused fragmentation is worse. Platforms that still only have a Linux Kernel framebuffer driver(fbdev) or no driver. Software that is not ARM64 compatible yet are now your worst problems.
    https://wiki.debian.org/Arm64Port
    There are a growing list of Arm64 boards that just work out the box with the stock Linux kernel.

    The big thing is a lot of things in Linux have to change to support the ARM world.

    We are use to a GPU being 1 bit. ARM world. what we call a GPU might be 3 independent parts Outputs, 2d rendering and 3d rendering all done by different blocks of hardware with direct controls and you might in fact have multiples of each. Nvidia is the one who has put forwards making a new framework to cope with this and all the ARM guys agree its needed.

    So the ARM hardware fragmentation problem was solve. Now the problem how to interface with the totally different design in a generic way problem has to be solved.

    Just to give you a idea how bad it is there are 300+ different vendors making opengl ES 3.2 compatible GPU units for ARM. This is not fragmentation but massive competition. The arm64 design means you can identify the GPUs, outputs and 2d rendering present…. Its now a problem of loading the right drivers and have them be able to work with each other without needing custom hacks.

    You think about for how long we have not been able to put ATI/AMD, Nvidia and Intel cards in a x86 computer and be able to use them all at once to their full ability. ARM platform mandates that complex mix of multi vendor video cards has to work or have limited platform coverage. Same barrier harms Android as well. How much of the Linux graphics stack was designed correctly for way ARM systems are design operate if you go back 2 years 0% and we would be barely at 5% now.

    Yes it simple to say it fragmentation but when the problem exists in x86 with ATI/AMD, Nvidia and Intel when using multi-able graphics cards is a Linux kernel and user-space design issue.

  150. DrLoser, losing in a big way, wrote, “He only desires ARM as a thin client (which it can just about manage) and as a server (which it cannot, as yet.)”

    Nonsense. Odroid-C2 is a decent thick client. Just ask TLW. Ask her to compare it to XP on x86 when it was new… Ask her about malware and slowing down and re-re-reboooting… Last time I checked, Odroid-C2 was ARMed. HP and others are selling ARMed servers and of course many others are buying them, so what’s with that? ARM does not have to be more powerful than x86 to be useful on desktop or server, just good enough, using less power and less expensive. It’s there.

  151. DrLoser says:

    You may wonder why I replied to Deaf Spy, Robert.
    Here’s the reason. With the sole exception of Dougie, who seems to know what he’s talking about, the rest of you are not worth the intellectual effort.
    You and Fifi in particular.

  152. DrLoser says:

    Unless someone gets the reigns, there will be no year of ARM on desktop, not next year, not the year after, but never.

    No fair, Deaf Spy. There will always be a place for ARM: traffic lights at one end, and mobile phones at the other end.
    Obviously not on the desktop. You are misrepresenting Robert’s fervent senile dreams here. He only desires ARM as a thin client (which it can just about manage) and as a server (which it cannot, as yet.)
    Robert has no interest in what the majority of the world uses for serious computing — ie, the desktop.
    I happened to trip over this interesting little article whilst thinking about the possible impending doom of Intel, btw.
    Apparently, and for no reason at all, investors don’t really believe Pog.
    Hard to understand, but there you go.

  153. dougman wrote, “Chromebooks are FAR more ubiquitous then any of the boards you have mentioned.”

    Does that matter, and does anyone care? The availability of ARM and */Linux on inexpensive boards is about choice and competition, not volume. If there is choice, millions will take it. Why would they not? GNU/Linux runs pretty well on TLW’s Odroid-C2. To her it’s as good as any PC except for a few small glitches like FF not running. She can use her old files. She can browse, edit text/images, search, play sound/video… It’s almost the same and the differences don’t matter much. Odroid-C2 was $40CDN and freight. We may buy a couple more for the other TVs in the home. This is mostly thanks to the boys and girls at Debian who keep a fleet of software ready for ARM. Then the folks who tinker with hardware found a way to boot Debian GNU/Linux and we’re off to the races. ARM on the desktop will keep getting better and it’s already good enough for many purposes.

  154. dougman says:

    “So, wishing it to go away is insufficient to stop it.”

    Wishing it go away? No one is saying or doing that, these devices pretty much do that themselves. Chromebooks are FAR more ubiquitous then any of the boards you have mentioned.

  155. Deaf Spy wrote, “Hardware ARM fragmentation is killing the platform the same way software Linux fragmentation did. Unless someone gets the reigns, there will be no year of ARM on desktop, not next year, not the year after, but never.”

    Nope. The world can and does make its own GNU/Linux desktops. 96boards.org/linaro have made some push into servers for the masses and RedHat and AMD have contributed to that. I grant you it is not high on their priorities but given the hardware the GNU/Linux desktop will thrive. The key is that several players need developers to have access to ARMed hardware with standards so there’s no way to stop the eventual availability of good ARMed hardware to the masses. Many developers are not willing to shell out $thousands for a board so the price will be reasonable. Further, the IoT and smartphone and tablet stuff is approaching the desktop from another direction. It’s only a matter of time before there’s enough memory and bandwidth to make all happy.

    An incident: Yesterday, while at dinner, a guest and I were in discussion about hunting. The guest was able to do everything I could do on a desktop from his smartphone to gather information without leaving his chair. The smartphones are very close to what a lot of desktop users need. Just add on-chip cache, RAM sockets, USB3, gigabit and SATA to a modified phone chip and you’re there.

    That’s what the Cello/Raspberry Pi and others are close to doing. The Odroid-C2 makes an excellent thin client and TLW is using it fairly well as a thick client. So, wishing it to go away is insufficient to stop it. Even now, one could lash up a cluster and get by. This year or next that won’t be necessary. All comers have something in the pipes that would be quite satisfactory.

    Check this out: A set-top box with guts similar to Odroid-C2 except the CPU is 8-core 2gHz… How long before such things are available like Odroid-C2? How long until they have way more RAM and SATA? I am disappointed by the slow pace of things but it’s happening nevertheless.

  156. Deaf Spy says:

    Dear, dear… You story, Robert, brings me back to the dark times when I was laboring on config.sys and autoexec.bat, juggling with IRQs and DMAs to get my non-IDE CD, and sound card, and mouse, and modem, all at the same time.

    Hardware ARM fragmentation is killing the platform the same way software Linux fragmentation did. Unless someone gets the reigns, there will be no year of ARM on desktop, not next year, not the year after, but never.

  157. oiaohm says:

    https://i.imgur.com/rmTaPvwl.jpg
    You do see some insane experiment on these small boards attempting to cool them way more than what they in fact need to be. That is ODROID XU-4 with a heatsink bigger than the board.

  158. oiaohm says:

    https://www.pugetsystems.com/mineral-oil-pc.php
    That is a wild solution to cool.

    https://www.quora.com/My-computer-CPU-constantly-runs-around-50-degrees-C-I-just-built-it-and-its-a-400-Intel-Should-I-get-a-water-cooler-for-it

    That’s rather hot.
    Dougman applying incorrect generalisation the question is with what nm is that is important. 28 nm chip in the ODROID C2. Does kinda make a difference 28 nm operates up to 90C. Safe is under 70C. Now 51C on an 14nm chip that is another matter 40-50C is safe for them so 51 is little too hot. Yes the move from 28nm to 14nm cuts the safe temperature by at least 20C. 51 is not that hot for a 28nm chip yet if it 51 C on current day 14nm intel chips it something to worry about. 10nm chips are showing the same temperature performance as 14nm. 7nm prototypes are about 45 C safe. 5nm and 3nm not enough chips yet but we can suspect closer to ambient temperatures.

    The new 1nm prototype transistor production tech might fall on the side of will not operate at general room ambient temperature.

    You still find 32nm chips made for space work and other places due to the higher temperature tolerance.

    Yes there could come a point where we need a peltier cooler to hold the silicon at operating temperature.

    Of course not knowing the topic as normal dogman goes and suggests something stupid.

  159. dougman says:

    “CPU temp: 51°C”

    That’s rather hot. You may want to place a tiny fan on the heat-sink powered off a thermocouple or better yet, place the device in the toilet and flush it a few times. The swirling movement of the water, will pull the heat away via convective heat transfer.

    At least, in a sense, you are not contributing to global warming and creating more hurricanes.

  160. Deaf Spy wrote, “Can the Odroid thing play 1080p video without stuttering?”

    Yes. That’s how TLW uses it all day long. However, that does use up almost all the juice. For normal browsing, she’s using about 30% of the CPU, which is just fine. It’s snappy, the way she likes it. The only remaining issues are that each distro, and I’ve tried four now, has one or more packages that just don’t work like FireFox segfaulting or VLC not working. These are things that can be fixed. Some of them are almost like regressions from years ago on x86 where FF kept waiting on stuff on slow systems and then the wait-loop ran out of resources. The issues are nails sticking up in the floor and will be pounded down with short words upstream. They apparently fixed the issues inadequately for ARMed system. ie. I think the issue is that their fixes didn’t allow for SD cards being so much slower than RAM or whatever. It’s not good to assume you’ll win a race in a wait-loop… You can spiral down the drain instead of climbing out of the hole. Midori seems to work very reliably except for video or lack of it. Not sure what that is about, but it could be packages missing. I can’t find any diagnostics and the FAQs don’t help much. Mplayer is just fine with video.
    ODROID C2

    Welcome to ARMBIAN Debian GNU/Linux 8 (jessie) 3.14.79-odroidc2
    System load: 0.56 Up time: 2 days Local users: 3
    Memory usage: 32 % of 1719Mb IP: 192.168.0.16
    CPU temp: 51°C
    Usage of /: 43% of 6.9G

  161. Deaf Spy says:

    I was replying to somebody who used the technical term, “Full HD”, which I have never heard before.

    Really? Do you leave in a cave, Robert?

    Well, let’s replace “Full HD”, which you’ve never heard of before, with 1080p.

    Can the Odroid thing play 1080p video without stuttering? Can it even play 720p at 25 fps?

  162. dougman wrote, “Now that this storm is “petering” out, the news media informed people they are dead if they stayed, was just wrong.”

    Were you so sure that the storm would not reach densely populated areas instead of thrashing the ocean? The winds at the eye were much stronger than what actually reached land and few homes would have survived let alone big box-like commercial buildings. Winds at the eye were ~140 mph. Winds that actually reached people were ~100 mph, damaging but not catastrophic. The death-toll and damage would have been huge if the hurricane had travelled on a track just 25 miles further west. There was no way to be certain of the path soon enough to allow people to think everything was going to be OK. It’s like a fire in a movie theatre. When it breaks out, you could choose to believe that it will soon be under control but the consequence of being wrong is not worth the risk. Further, these folks are going to assume they will be right next time too, when “the big one” actually rips them up. These folks may have proper building-codes but they don’t have storm shelters in the back yard like many in “tornado alley” and many don’t have basements or “strong rooms”. With 140 mph winds, there would have been much more devastation than these folks could “ride out”. As it is, the damage will amount to a huge sum. Being lucky is not at all the same as being smart.

  163. dougman says:

    Now that this storm is “petering” out, the news media informed people they are dead if they stayed, was just wrong.

    For instance, “Hurricane Matthew ‘will kill you,’ says Florida governor” the actual death tolls thus far in Florida are a total of four, two from trees and two from medical reasons.

    Then you have some hurricane expert’s chilling warning “this is not hype, this storm will make history”

    Yes, the only thing this storm did is benefited the gas stations and Home Depots greatly.

  164. dougman says:

    The eye-wall of the hurricane is well defined it seems, but weakening.

    Hurricane Matthew – Amateur Satellite Snapshot

  165. dougman says:

    “Well, it’s been more than a century since a hurricane so large came so close…”

    Say what? So are you saying, there hasn’t been any storms that made landfall in Florida, since Matthew?

    “What happened since then?”

    Nothing has changed to be honest.

    “A few millennia of carbon was burned.”

    As physics guy, surely you understand chemical reactions and know that carbon-dioxide does not burn; this is the supposed greenhouse gas that everyone complains about. Ignorantly, these same people cannot comprehend that vegetation requires carbon-dioxide, and that humans release it when breathing out.

    “Face it. A few miles to the west would have made things quite different.”

    Lol..a few miles west? Hurricanes are not that localized. Some of the reports I have read were stating just heavy winds, rain, some minor flooding and sporadic power outages.

  166. dougman wrote, “linking hurricane Matthew to climate change is overblown hype”.

    Well, it’s been more than a century since a hurricane so large came so close… What happened since then? A few millennia of carbon was burned. Face it. A few miles to the west would have made things quite different.

    Relatives in FL: They stayed up all night partying with friends and family so they would all live or die together. They are tired but otherwise OK.

  167. dougman says:

    Speaking of amazement, linking hurricane Matthew to climate change is overblown hype!

    I have been through numerous hurricanes, namely Sandy, Irene and Isabel while living near or on the water. Lots of wind, rain and flooding. This is why I have a 10KW generator with adequate supplies of stored food.

  168. The Wiz wrote, “4K is actually referred to as UHD”.

    I was replying to somebody who used the technical term, “Full HD”, which I have never heard before.

  169. The Wiz wrote, “the experiences that you have recounted in setting up the the Odrioid C2 Have done more to demonstrate its unsuitability in a world peopled by consumers who have zero interest in or ability to cope with such a hackers delight.

    I don’t think I’ve ever claimed that burning images onto SD-cards and such was consumer-oriented. I’ve claimed the hardware is sufficiently powerful to please consumers. That’s a fact and similar hardware is shipped to consumers as set-top boxes and such with Android/Linux or just some fixed menu. TLW is quite able to do her tasks with Odroid-C2 but I had to set it up with her file-shares and applications and menus…

    NB – Sent from the midori web-browser which I’m running on Odroid-C2 at the moment because FF segfaults…

  170. Wizard Emeritus says:

    “Some folks call 4K HD. Some call 720p HD. I’ve seen them all. Odroid-C2 is using 1080p at the moment. It’s nice.”

    4K is actually referred to as UHD TV Robert Pogson. 4K UHD is also rapidly becoming the new high end standard. AFIAK the technology behind the Odroid C2 is not going to be able to handle this, so its desirability for anyone beyond a cheapskate like you will decline. Then again, at $40 it is basically a throwaway.

    The real point of my comment, however, is that the experiences that you have recounted in setting up the the Odrioid C2 Have done more to demonstrate its unsuitability in a world peopled by consumers who have zero interest in or ability to cope with such a hackers delight.

  171. Deaf Spy wrote, “did you incidentally forgot to use Google?”

    Some folks call 4K HD. Some call 720p HD. I’ve seen them all. Odroid-C2 is using 1080p at the moment. It’s nice.

  172. Deaf Spy says:

    Define Full HD
    Well, did you incidentally forgot to use Google? Anyway. Full HD – video mode with 1920 x 1080 pixels resolution. You know, crispy, nice looking videos.

  173. Deaf Spy wrote, “Odroid C2 still can’t play full HD videos, driver updates will never come, and it is October, and the world is not nuts about ARM.”

    Define Full HD. TLW is satisfied with the little guy. He just keeps working and working and working…

  174. Deaf Spy says:

    I sense hate in you tonight, whats wrong?
    Odroid C2 still can’t play full HD videos, driver updates will never come, and it is October, and the world is not nuts about ARM.

  175. dougman says:

    “Will the only good to come out of these events be that a million Trump-supporters have died, that the gene-pool will have been sharpened for rationality traits, or that the economy will have a major boost repairing the damage?”

    That is a shallow thing for you to say Robert. What would TLW say if a cyclone of immense proportions hit’s the Philippines? Even if I share a political discordance with someone, I don’t wish them dead, that’s just wrong.

    You may set back in stare in bewilderment, but Florida is use to hurricanes. Lot’s of people will stay at home sticking this one out, namely as Florida is on the “dry-side” of the storm. Lots of wind and bands of rain, but not the normal soaker one experiences.

    I sense hate in you tonight, whats wrong?

Leave a Reply